The Ecology and Population structure of the invasive ... · aufgefunden (mehrheitlich neue, unbeschriebene Arten), von denen die Grille Myrmecophilus pallidithorax eingehender untersucht

The Ecology and Population structure

of the invasive Yellow Crazy Ant

Anoplolepis gracilipes

Dissertation zur Erlangung des

naturwissenschaftlichen Doktorgrades der

Julius-Maximillians-Universität Würzburg

vorgelegt von

Jochen Drescher

aus Köln

Würzburg 2011

ii

Eingereicht am:

Mitglieder des Promotionskolloqiums

Vorsitzender:

Gutachter: PD Dr. Nico Blüthgen, Julius-Maximillians-Universität Würzburg

Gutachter: PD Dr. Thomas Schmitt, Albert-Ludwigs-Universität Freiburg

Tag des Promotionskolloqiums:

Doktorurkunde ausgehändigt am:

iii

Ants are everywhere, but only occasionally noticed. They run much of the terrestrial world as

the premier soil turners, channelers of energy, dominatrices of the insect fauna – yet receive

only passing mention in textbooks on ecology. [...] The neglect of ants in science and natural

history is a shortcoming that should be remedied, for they represent the culmination of insect

evolution, in the same sense that human beings represent the summit of vertebrate evolution.

Bert Hölldobler & Edward O. Wilson, ‘The Ants’, 1990

iv

v

ERKLÄRUNGEN

gemäß §4, Abs. 3, S. 3, 5 und 8 der Promotionsordnung vom 15. März 1999

(zuletzt geändert durch Satzung vom 12. August 2009)

Hiermit erkläre ich ehrenwörtlich, daß ich die vorliegende Dissertation selbständig angefertigt

und keine anderen als die von mir angegebenen Quellen und Hilfsmittel benutzt habe.

Die Dissertation hat weder in gleicher noch in ähnlicher Form einem anderen

Prüfungsverfahren vorgelegen.

Am 04. September 2006 hat mir die Universität Würzburg den akademischen Grad „Diplom-

Biologe Univ.“ verliehen. Weitere akademische Grade habe ich weder erworben noch

versucht zu erwerben.

Würzburg, 18. März 2011

__________________________

Jochen Drescher

vi

This thesis is based on the following manuscripts:

Drescher J, Blüthgen N, Feldhaar H (in prep) Mitochondrial DNA diversity of the Yellow

Crazy Ant Anoplolepis gracilipes in Southeast Asia and the Indopacific

Drescher J, Blüthgen N, Schmitt T, Bühler J, Feldhaar H (2010) Societies drifting apart?

Behavioural, genetic and chemical differentiation between supercolonies in the

Yellow Crazy Ant Anoplolepis gracilipes. PLoS ONE 5(10): e13581

Drescher J, Blüthgen N, Schmitt T, Bühler J, Babis M, Feldhaar H (in prep) Worker

aggression towards allocolonial reproductives facilitates speciation between

supercolonies in the Yellow Crazy Ant Anoplolepis gracilipes

Drescher J, Feldhaar H, Blüthgen N (2011) Interspecific aggression and resource

monopolization of the invasive ant Anoplolepis gracilipes in Malaysian Borneo.

Biotropica 43: 93-99

Drescher J, Rinke S, Schmitt T, Feldhaar H, Blüthgen N (in prep) Catch me if you can:

Agility of the ant guest Myrmecophilus pallidithorax as a major factor determining

survival in colonies of the Yellow Crazy Ant Anoplolepis gracilipes

vii

TABLE OF CONTENTS

I. ZUSAMMENFASSUNG 1

II. SUMMARY 5

III. GENERAL INTRODUCTION 9

IV. SYNOPSIS 19

V. MITOCHONDRIAL DNA DIVERSITY OF THE 25

YELLOW CRAZY ANT ANOPLOLEPIS GRACILIPES

IN SOUTHEAST ASIA AND THE INDOPACIFIC

VI. GENETIC AND CHEMICAL DIFFERENTIATION 37

BETWEEN ANOPLOLEPIS GRACILIPES SUPERCOLONIES

VII. WORKER AGGRESSION TOWARDS ALLOCOLONIAL 57

MALES AND QUEENS MAY FACILITATE SPECIATION

BETWEEN ANOPLOLEPIS GRACILIPES SUPERCOLONIES

VIII. ECOLOGICAL DOMINANCE OF ANOPLOLEPIS GRACILIPES 73

IN AN ANT COMMUNITY IN NORTH-EAST BORNEO

IX. AGILITY DETERMINES THE SURVIVAL OF THE CLEPTO- 89

PARASITIC CRICKET MYRMECOPHILUS PALLIDITHORAX

IN SUPERCOLONIES OF ANOPLOLEPIS GRACILIPES

X. REFERENCES 101

XI. ACKNOWLEDGEMENTS 115

XII. CURRICULUM VITAE 117

viii

I. Zusammenfassung

1

I. ZUSAMMENFASSUNG

Die anthropogene Verschleppung und anschließende Etablierung invasiver Arten in neuen

Habitaten ist einer der Hauptfaktoren des globalen Rückganges biologischer Vielfalt. Invasive

Ameisen gehören zu den besonders schädlichen Invasoren und stellen fünf der 100 weltweit

problematischsten invasiven Organismen. Ungeachtet einer Vielzahl von Studien über lokale

ökologische Konsequenzen invasiver Ameisen sind die Mechanismen, die den Invasionen

zugrunde liegen, kaum verstanden und weitestgehend auf die zwei Ameisenarten Linepithema

humile und Solenopsis invicta beschränkt. Das Hauptziel der vorliegenden Dissertation war

durch Untersuchung der ebenfalls invasiven Ameisenart Anoplolepis gracilipes zu einem

besseren Verständnis der Mechanismen biologischer Invasionen und deren ökologischen

Konsequenzen beizutragen.

Anoplolepis gracilipes ist eine in den Tropen weit verbreitete invasive Ameisenart, die in

anthropogen gestörten Habitaten Südostasiens und des indopazifischen Raumes sehr häufig

vorzufinden ist. Während detaillierte Informationen bezüglich ihres derzeitigen

Verbreitungsgebietes vorliegen, ist ihre geographische Herkunft immer noch unbekannt.

Weiterhin ist unklar, in welchem Maße die Sozialstruktur von A. gracilipes zu ihrer

ökologischen Dominanz beiträgt und wie sich diese wiederum in einem potentiellen

Herkunftsgebiet (Südostasien) darstellt.

Mitochondriale DNA-Sequenzen legen nahe, dass die überwiegende Mehrheit der im

indopazifischen Raum vorkommenden Kolonien von südostasiatischen Populationen

eingeführt wurde. Die südostasiatischen Kolonien entstammen möglicherweise einem bislang

unbekanntem Ursprungsgebiet.

Verhaltenstests und genetische Analysen ergaben, dass Superkolonien von A. gracilipes aus

sehr nah verwandten Individuen bestehen, womit sie stark vergrößerten Varianten

monogyner, polydomer Kolonien anderer Ameisenarten ähneln. Ausserdem wiesen sowohl

genetische Daten sowie Profile epikutikulärer Kohlenwasserstoffe auf eine erhebliche

Differenzierung zwischen verschiedenen Superkolonien hin. Das Ausmaß der genetischen

und chemischen Differenzierung deutet darauf hin, dass Genfluss zwischen Superkolonien

stark reduziert oder sogar unterbrochen ist. Da die Paarung bei A. gracilipes wahrscheinlich

nur im eigenen Nest stattfindet (Hochzeitsflüge wurden bislang noch nicht beobachtet),

könnte eine positive Rückkopplung zwischen Aggression, Verwandtschaftsgrad und

I. Zusammenfassung

2

epikutikulärer Chemie dazu führen, dass die Differenzierung zwischen Superkolonien durch

eine Kombination aus genetischer Drift und neutraler Evolution weiter verstärkt wird.

Superkolonien, die nicht durch Genfluss miteinander im Austausch stehen, könnten sich also

konsequenterweise in unterschiedliche evolutive Richtungen entwickeln. Eine der

Möglichkeiten, durch die Genfluss zwischen verschiedenen Superkolonien aufrecht erhalten

werden könnte, wäre deshalb die Einwanderung reproduktiver Individuen (Königinnen und

Männchen) in fremde Superkolonien. Meine Untersuchungen ergaben, dass die Migration von

Männchen und Königinnen zwische verschiedenen Superkolonien jedoch durch die

Arbeiterinnen unterbunden wird, welche in erhöhten Maße aggressiv gegenüber

Geschlechtstieren anderer Superkolonien waren. Weiterhin deuteten Kreuzungsexperimente

zwischen koloniefremden Männchen und Königinnen darauf hin, dass Superkolonien von A.

gracilipes unter Umständen schon reproduktiv isoliert sind, welches konsequenterweise zur

Diversifizierung von A. gracilipes in verschiedene Arten führen sollte.

Bezüglich ihrer potentiellen ökologischen Dominanz in Nordost-Borneo konnte gezeigt

werden, dass A. gracilipes die lokale Ameisenfauna erheblichen beeinflusst. Innerhalb der

Superkolonien von A. gracilipes fanden sich sowohl weniger Ameisenarten als auch eine

andere Artzusammensetzung als ausserhalb. Die Ergebnisse deuteten darauf hin, dass die

ökologische Dominanz von A. gracilipes maßgeblich auf der Monopolisierung von

Nahrungsquellen beruht. Diese wird ermöglicht durch eine Kombination aus schneller

Rekrutierung von Nestgenossinnen, zahlenmäßiger Überlegenheit und ausgeprägter

interspezifischer Aggression. A. gracilipes kommt fast ausschließlich in anthropogen

gestörten Habitaten wie Wohngebieten oder landwirtschaftlich genutzten Flächen vor. Die

zunehmende Habitatkonversion in Nordost-Borneo führt zu einem enormen Anstieg der von

A. gracilipes besiedelbaren Habitate, so dass mit einem signifikanten Populationswachstum

von A. gracilipes zu rechnen sein wird. Ein schnelles Populationswachstum sowie

ökologische Dominanz sind jedoch nicht allein auf invasive Arten geprägte Charakteristika,

sondern können auch bei nativen Arten zu beobachten sein, welche durch zunehmende

Verfügbarkeit anthropogen veränderten Habitats zu Schädlingen werden können.

Abschließend wurden mehrere Arten potentieller Sozialparasiten in Nestern von A. gracilipes

aufgefunden (mehrheitlich neue, unbeschriebene Arten), von denen die Grille Myrmecophilus

pallidithorax eingehender untersucht wurde. Verhaltenstests und die Analyse kutikulärer

Kohlenwasserstoffe zeigten, dass M. pallidithorax von ihrem Wirt bei jeder sich bietender

I. Zusammenfassung

3

Gelegenheit angegriffen und sogar verzehrt wird. Jedoch kann sie den Aggressionen ihres

Wirtes weitestgehend ausweichen dank schneller Fluchtreflexe sowie, möglicherweise,

chemischer Tarnung.

Die vorliegende Dissertation zeigt, dass lang zurückliegende Invasionen die Unterscheidung

zwischen eingeführten oder nativen Schädlingen erschweren, da beide tiefgreifende

ökologische Einflüsse auf native Artengemeinschaften haben können. Es wurde weiterhin

deutlich, dass die außergewöhnliche Sozialstruktur von invasiven Ameisen wie A. gracilipes

ihre ökologische Dominanz begründet. Die Bildung von Superkolonien bei invasiven

Ameisen stellt zudem nicht notwendigerweise eine evolutive Sackgasse dar, sondern kann im

Gegenzug sogar zur Artbildung führen, begünstigt durch ungewöhnliche Paarungs- und

Verbreitungsstrategien. Abschließend zeigte sich, daß die Untersuchung von Sozialparasiten

invasiver Ameisen unter Umständen Rückschlüsse erlaubt bezüglich des historischen

Ablaufes einer Invasion, der Biogeographie des invasiven Organismus sowie potentiellen

Gegenmaßnahmen gegen die beteiligten Invasoren.

I. Zusammenfassung

4

II. Summary

5

II. SUMMARY

The anthropogenic introduction and subsequent establishment of invasive alien species is a

major factor contributing to decline of biodiversity in ecosystems worldwide. Invasive ants

are particularly damaging invaders, constituting five of the 100 world’s worst invasive alien

organisms. While there are numerous reports describing the local ecological consequences of

ant invasions are abundant, general insights on the underlying mechanisms are scarce and

largely restricted to two ant species, i.e. the Argentine Ant Linepithema humile and the Red

Imported Fire Ant Solenopsis invicta. The primary objective of this thesis was to broaden the

spectrum of studied invasive ant species in order to facilitate a more general understanding of

the mechanisms and ecological consequences of ant invasions.

The invasive Yellow Crazy Ant Anoplolepis gracilipes is a widespread tropical ant species

which is particularly common in anthropogenically disturbed habitats in South-East Asia and

the Indopacific region. Despite detailed information on its current distribution, its native range

is unknown, and there is little information concerning its social structure as a potential

mechanism facilitating invasion as well as its ecology in one of the putative native ranges,

South-East Asia.

Using mitochondrial DNA sequences (mtDNA haplotypes), I demonstrated that the majority

of the current Indopacific colonies were likely introduced from South-East Asian populations,

which in turn may have been introduced much earlier from a yet unidentified native range.

By conducting behavioural, genetic and chemical analyses, I found that A. gracilipes

supercolonies contain closely related individuals, thus genetically resembling enlarged

versions of monogynous, polydomous colonies of other ant species. Furthermore, the data

showed that mutually aggressive A. gracilipes supercolonies were highly differentiated both

genetically and chemically, suggesting limited or even absent gene flow between

supercolonies. Intranidal mating and colony-budding are most likely the predominant, if not

the exclusive mode of reproduction and dispersal strategy of A. gracilipes. Consequently, a

positive feedback between genetic, chemical and behavioural traits may further enhance

supercolony differentiation though genetic drift and neutral evolution. This potential scenario

led to the hypothesis that absent gene flow between different A. gracilipes supercolonies may

drive them towards different evolutionary pathways, possibly including speciation. Thus, I

examined one potential way by which gene flow between supercolonies of an ant species

II. Summary

6

without nuptial flights may be maintained, i.e. the immigration of sexuals into foreign

supercolonies. The results suggest that this option of maintaining gene flow between different

supercolonies is likely impaired by severe aggression of workers towards allocolonial sexuals.

Moreover, breeding experiments involving males and queens from different supercolonies

suggest that different A. gracilipes supercolonies may already be on the verge of reproductive

isolation, which might ultimately lead to the diversification of A. gracilipes into different

species.

Regarding the ecological consequences of its potential introduction to NE-Borneo, I could

show that A. gracilipes supercolonies may seriously affect the local ant fauna. The ant

community within supercolonies was significantly less diverse and differed in species

composition from areas outside supercolonies. My data suggest that the ecological dominance

of A. gracilipes within local ant communities was facilitated by monopolization of food

sources within its supercolony territory, achieved by a combination of rapid recruitment,

numerical dominance and pronounced interspecific aggression. A. gracilipes’ distribution is

almost exclusively limited to anthropogenically altered habitat, such as residential and

agricultural areas. The rate at which habitat conversion takes place in NE-Borneo will provide

A. gracilipes with a rapidly increasing abundance of suitable habitats, thus potentially

entailing significant population growth. An potentially increasing population size and

ecological dominance, however, are not features that are limited to invasive alien species, but

may also occur in native species that become ‘pests’ in an increasing abundance of

anthropogenically altered habitat.

Lastly, I repeatedly detected several ant guests (most of them being new species) in

supercolonies of A. gracilipes. I subsequently describe the relationship between one of them

(the cricket Myrmecophilus pallidithorax) and its ant host. By conducting behavioural

bioassays and analyses of cuticular hydrocarbon (CHC) profiles, I revealed that although M.

pallidithorax is ferociously attacked and even consumed by A. gracilipes whenever possible,

it may evade aggression from its host by a combination of supreme agility and, possibly,

chemical deception.

Overall, this thesis adds to our general understanding of biological invasions by contributing

species-specific data on a previously understudied invasive organism, the Yellow Crazy Ant

Anoplolepis gracilipes. It demonstrates that introductions which may have occurred a

considerably long time ago may make it difficult to determine whether a given species is an

II. Summary

7

introduced invader or a native pest species, as both may have pronounced ecological effects in

native species communities. Furthermore, this thesis suggests that supercolonialism in

invasive ants may not necessarily be an evolutionary dead end, but that it may possibly give

rise to new species due to reproductive boundaries between supercolonies evoked by peculiar

mating and dispersal strategies. Lastly, it shows that invasive species may be subject to

parasitism, either by native or likewise introduced species, and that the nature of these

interspecific associations may hold important information concerning invasion history,

biogeography or even countermeasures against the respective alien invader.

II. Summary

8

III. General Introduction

9

III. GENERAL INTRODUCTION

The increasing human population with its growing demands for resources is globally altering

natural ecosystems. Many of these changes are purposeful and intended to benefit society,

such as agriculture or stock farming. Others, albeit intentional, have negative side effects that

may threaten the goods and services delivered to society by natural ecosystems (Mooney

2005). These benefits to humankind include provisioning services (e.g. food, water,

pharmaceuticals, energy), regulating services (e.g. disease, climate, decomposition,

pollination) and supporting services (e.g. nutrient dispersal and circulation, seed dispersal,

soil formation, primary production) (Millennium Ecosystem Assessment, MA, 2003). Many

ecosystem services are positively associated to biodiversity, i.e. high levels of biodiversity are

linked to reliable delivery of ecosystem services and vice versa (Worm et al. 2006; Nelson et

al. 2009). Biodiversity in both terrestrial and marine ecosystems, however, is in worldwide

decline (Whitmore and Sayer 1992; Jones et al. 2004; Sodhi et al. 2004; Clausen and York

2008), thus threatening the supply of ecosystem services to human society. There are several

reasons for the massive decline of global biodiversity (see Swanson 1995) and among the

most important threats ranks the establishment of invasive alien species (Mack et al. 2000a;

Sala et al. 2000; Mooney and Cleland 2001).

Biological Invasions

Invasive organisms are found in a vast variety of taxa, including marine and terrestrial

vertebrates and invertebrates, plants and fungi. The term ‘biological invasion’ usually refers

to the ecologically harmful establishment of anthropogenically introduced organisms,

regardless whether the organisms were introduced intentionally or accidentally. An important

part of this definition is that alien species have to cause ecological damages to be termed

‘invasive’, as not all introduced organisms become established and not all established

intruders become ecological problems (Williamson 1996). However, given the high frequency

by which alien organisms are introduced into ecosystems worldwide, the number of species

that become established and cause ecological problems can be quite large. For instance, of the

ca. 1500 exotic insects that have become part of the U.S. insect fauna over the past 20 years,

16% are regarded invasive pests (i.e. 235 species, Pimentel 1993). The observation that only a

fraction of introduced species turn out to be ecologically harmful (thus invasive) has led

Williamson and Fritter (1996) to propose the ‘tens rule’, by which 10% of the introduced

species become established, of which 10% become invasive. While some reports support the


10

predictions made by the tens rule (e.g. Pimentel 1993; Boudouresque and Verlaque 2002),

others show that the proportion of established species that become invasive can be much

greater, often exceeding 50% (Kiritani and Yamamura 2003; Jeschke and Strayer 2005).

One of the most-cited hypotheses explaining the higher success of invasive species compared

to native species is the ‘enemy-release hypothesis’ (‘ERH’) (Keane and Crawley 2002;

Colautti et al. 2004; Liu and Stiling 2006; Liu et al. 2006). The ERH states that the success of

an invading organism – often measured as abundance, population density or capacity to

displace native species – is due to the scarcity of natural enemies (pathogens, parasites,

predators) in its introduced range compared to its native range. Since its original proposition

(Williamson 1996; Crawley 1997), various experimental studies have revealed supportive

evidence verifying the ERH (Mitchell and Power 2003; Torchin et al. 2003; White et al. 2008;

Adams et al. 2009; Cincotta et al. 2009). However, using the ERH as the sole explanation

biological invasions is inappropriate, as a multitude of other factors may contribute to the

success of individual invasive alien species (Colautti et al. 2004).

Just like invasive species, native organisms can also have negative ecological and economical

impacts and are predominantly referred to as ‘pests’ (as opposed to ‘invasive alien species’).

In the USA, approx. a third of the major crop weeds, about half of the pasture weeds and three

quarters of the major forest pests are from within the country (Pimentel 1993; Mooney 2005).

However, invasive alien species appear to be cause more serious problems than native

species, at least in terms of weeds (Pimentel et al. 2005).

Consequences of biological invasions

Biological invasions may entail severe ecological consequences, ranging from increased

resource competition for native species, regional displacement and even extirpation of

individual native species to the alteration of entire ecosystems (reviewed by Crooks 2002).

Well-known examples of ecologically devastating invaders include the Brown Tree Snake

Boiga irregularis, which is held responsible for the extinction of 13 of the 22 native breeding

birds as well as several bat and lizard species on Guam (Savidge 1987; Rodda and Savidge

2007), or the Nile perch Lates niloticus which contributed to the demise of dozens of cichlid

species in the Great East-African Lakes (Witte et al. 2000; Verschuren et al. 2002; Aloo

2003). As a consequence, invasive alien species are often regarded as belonging to the most

important drivers of the globally progressing environmental change (Vitousek et al. 1996;


11

Mack et al. 2000a; Sala et al. 2000; Mooney et al. 2005; Perrings et al. 2010). Moreover,

invasive alien species often cause severe economic damage (Perrings et al. 2002; Born et al.

2005; Pimentel et al. 2005), either indirectly by the disruption of ecological processes (and

thus a lowered supply of ecosystem services, see above) or directly by the infestation of

agricultural facilities or other commercial structures. For instance, the costs for cleaning

intake pipes of US-American waterworks from colonies of the invasive European zebra

mussel Dreissena polymorpha have amounted to approx. 3.1 billion USD between 1993 and

2003 (Vitousek et al. 1996), and the combined costs for its economic damage and eradication

measures are currently estimated to be as high as one billion USD per year (Pimentel et al.

2005). Overall, Pimentel et al. (2005) estimated the costs associated with the economic

damages and control measures of more than 50,000 invasive alien species in the USA to be as

high as 120 billion USD per year (Pimentel et al. 2005).

Vectors of biological invasions and susceptibility to invasion

There are various means by which invasive organisms reach previously uninhabited

ecosystems (via ‘invasion vectors’). The vast majority of motorized vehicles used in both

private and commercial traffic today (i.e. ships, planes, automobiles) act as invasion vectors,

introducing up to dozens of alien species into foreign habitats in one go (reviewed by Ruiz

and Carlton 2003). Terrestrial ecosystems that evolved in isolation, constrained by

biogeographic barriers such as oceans or mountain ranges, have now become functionally

connected due to the global human traffic which moves biological material biological material

over long distances in relatively short amounts of time (Mooney 2005). Similarly, coastal

ecosystems of different continents, though likewise separated by oceans, have become linked

by the increasing amount of intercontinental cargo ships carrying a multitude of potentially

invasive species and exotic pathogens in their ballast water tanks (Ruiz et al. 2000; Baier et al.

2003; Drake and Lodge 2004; Drake et al. 2007).

The vulnerability of ecosystems to this global exchange of biota, however, may differ

between regions, with islands seeming particularly prone to biotic invasions (Loope and

Mueller-Dombois 1989; Atkinson and Cameron 1993; McDonald and Cooper 1995). The

susceptibility of island ecosystems to biotic invasions is reflected by the large proportion of

non-indigenous species inhabiting their flora and fauna, e.g. up to 65% of all the vascular

plant species found on Bermuda (Vitousek et al. 1996), nine out of 20 species in the Auckland

islands avifauna (Case 1996) or more than 20% of all the biota on Hawaii (across various


12

taxa, Eldredge and Miller 1997). The reasons for the higher susceptibility of islands to biotic

invasions relative to comparable mainland areas has been subject to much debate, and is often

attributed to the differences in biotic communities between islands and continents. Compared

to continental ecosystems, biological islands usually support communities of low diversity

and high differentiation (Sax and Brown 2000) which are often unsaturated (Moulton and

Pimm 1986) and thus may contain wide or open ecological niches (e.g. Olesen et al. 2002).

Diversity, in turn, may facilitate biotic resistance to invasion (Tilman 1997; Stachowicz et al.

2002), thus possibly explaining the lower invasion rates observed on continents compared to

insular ecosystems.

Biological invasions by ants

Social insects, especially ants, rank among the most devastating groups of invaders (Moller

1996). More than 12,500 ant species have been described to date (http://www.antbase.org/;

assessed in March 2011) of which approx. 150 have been introduced into new habitats by

humans (McGlynn 1999a, b). The majority of introduced ant species remain confined to

anthropogenically disturbed habitats. Several introduced ant species, however, have the

capacity to invade natural ecosystems and are thus termed ‘invasive’ (Holway et al. 2002).

The Global Invasive Species Database (GISD, managed by the Invasive Species Specialist

Group ISSG of the IUCN Species Survival Commission) currently lists 19 ant species as

invasive, five of which have been rated as belonging to the 100 worst invasive species (Lowe

et al. 2000). These five ant species are the Yellow Crazy Ant Anoplolepis gracilipes, the

Argentine Ant Linepithema humile, the Big-Headed Ant Pheidole megacephala, the Red

Imported Fire Ant Solenopsis invicta and the Little Fire Ant Wasmannia auropunctata. Until

recently, insights on the causes and consequences of ant invasions were mainly derived from

studies on two focal species, i.e. L. humile and S. invicta (Holway et al. 2002). Like biological

invasions by other taxa, successful establishment of invasive ant populations involve both

economic and ecological damages. For instance, the establishment of S. invicta in the USA is

estimated to cause crop losses and general damages amounting to 600 million USD p.a.,

followed by 400 million UDS p.a. for control and eradication measures (Pimentel et al. 2005).

Ecologically, ant invasions have repeatedly been shown to entail significant decreases in

diversity and abundance of both vertebrate and invertebrate populations (Holway 1998;

Suarez and Case 2002; Suarez et al. 2005; Wetterer and Moore 2005; Davis et al. 2008;

Hoffmann and Parr 2008). Occasionally, invasive ants may even alter the structure of entire


13

ecosystems, e.g. by displacement of key-stone species (O'Dowd et al. 2003), competitive

advantages over native invertebrates (McNatty et al. 2009), reduction of fruit and seed

production of native plants (Lach et al. 2010), or by disruption of essential mutualisms such

as seed dispersal (Christian 2001; Davis et al. 2010) or pollination (Blancafort and Gomez

2005). Moreover, invasive ants regularly affect native ant communities (Holway 1999;

Human and Gordon 1999; Sanders et al. 2003; Lessard et al. 2009), which may lead to severe

community-level changes given the diversity of roles that ants play in terrestrial ecosystems

(Hölldobler and Wilson 1990; Holway et al. 2002). Invasive ants share several characteristics

that facilitate their superiority over native ant species, both in terms of direct competition as

well as indirect competition for resources. Among these features are pronounced interspecific

aggression (Holway 1999; Human and Gordon 1999), superiority at resource discovery and

monopolization (Davidson 1998; Holway 1999) and numerical dominance within the invaded

habitat (Walters and Mackay 2005). Furthermore, the majority of invasive ant species forms

large, polygynous and polydomous ‘supercolonies’ (see below), in which males and queens

often mate within the nest (‘intranidal mating’) and which grow by ‘budding’, i.e. the

establishment of additional nests sites at the supercolony fringe by supercolony subsets

(‘propagules’) containing one or several queens, brood and workers (reviewed by Holway et

al. 2002; Helanterä et al. 2009). This social structure may likely contribute to the ecological

success of invasive ant species through positive feedback, i.e. the larger a supercolony gets,

the more likely it is to outcompete other ant species and thus grow further (e.g. Helanterä et

al. 2009). And finally, invasive ants may have a further advantage over native ant species in

that they have repeatedly been shown to possess unusual reproductive systems, e.g. clonal

reproduction of males and queens in the invasive longhorn ant Paratrechina longicornis

(Pearcy et al. 2011) and little fire ant Wasmannia auropunctata (Fournier et al. 2005; Foucaud

et al. 2009), which may further contribute to increased supercolony growth. All of these

characteristics are, to a certain degree, connected to the special population structure that is

almost exclusively limited to invasive ant species.

Population structure of invasive ant species

Most ant species show a multicolonial population structure, in which a population is

subdivided into individual colonies which aggressively compete for resources such as

territory and food. Such colonies are closed family units, usually containing one egg-laying

queen and her often sterile daughter workers. Some ant species, however, show a unicolonial


14

population structure, where local populations are characterized by weak or nonexistent colony

boundaries and in which intraspecific aggression is reduced or absent (Wilson 1971;

Hölldobler and Wilson 1977). Local populations of intraspecifically tolerant colonies are

referred to as supercolonies (Giraud et al. 2002; Buczkowski et al. 2004; Jaquiery et al. 2005),

where aggression between nests is absent and workers, queens, brood and food items are

exchanged freely between nests. While the same may apply to polygynous (more than one

queen) and polydomous colonies (more than one nest) of ‘normal’ ant species, established

supercolonies contain “[…] such a large number of nests that direct cooperative actions are

impossible between individuals in distant nests” (Pedersen et al. 2006). Invasive ant species

have repeatedly been reported to form polydomous and polygynous supercolonies in their

introduced range (Suarez et al. 2002; Espadaler et al. 2004; Errard et al. 2005; Pedersen et al.

2006; Thomas et al. 2006; Cremer et al. 2008; Ugelvig et al. 2008; Thomas et al. 2010a) and

sometimes even in their native ranges (Pedersen et al. 2006; Vogel et al. 2009). Supercolonies

of invasive ant species may be extremely large, as demonstrated by the main European

supercolony of L. humile which covers thousands of kilometres along the coast of the Iberian

Peninsular (Giraud et al. 2002; Jaquiery et al. 2005) and which may even span across several

continents (Brandt et al. 2009b). While different supercolonies exhibit aggression between

each other, the lack of aggression between nests from the same supercolony may be

energetically advantageous, as costs associated with aggression are reduced (Reeve 1989;

Steiner et al. 2007). This, in turn, may allow invasive ant supercolonies to sustain worker

densities exceeding those of native ants, thus potentially entailing numerical superiority

(Holway et al. 1998; Giraud et al. 2002; Suarez et al. 2008) and ecological dominance over

native ants within the invaded habitat (Holway 1999; Morrison 2000; Holway et al. 2002).

Unicolonial populations and supercolonies have been reported from both invasive and non-

invasive ant species (reviewed by Helanterä et al. 2009) as well as termites (Leniaud et al.

2009; Perdereau et al. 2010; Vargo and Husseneder 2011). The origins of supercolonialism

(i.e. the ability to form supercolonies), however, are not entirely clear and several mechanisms

have been proposed (e.g. Keller and Ross 1999; Tsutsui et al. 2000; Giraud et al. 2002;

Steiner et al. 2007). In the Red Imported Fire Ant Solenopsis invicta, the formation of

supercolonies seems to be exclusively linked to a single locus, General Protein Gp-9 (Keller

and Ross 1999). Colonies of S. invicta can be either monogynous or polygynous (Glancey et al.

1987) according to the alleles that queens bear on the Gp-9 locus (Keller and Ross 1999). Gp-

9 affects the expression of a fertility signal, which in turn affects aggression between S.


15

invicta colonies (Krieger 2005). Queens with a homozygote dominant genotype on this locus

are independent colony foundresses, while queens with a heterozygote genotype jointly form

polygynous colonies which are referred to as supercolonies (Macom and Porter 1996; Vander

Meer and Porter 2001). Polygynous S. invicta supercolonies may be up to two times larger

than monogyne colonies (Balas and Adams 1996; Macom and Porter 1996) and food is

exchanged between different nests of the same supercolony, depending on the distance

between nests, the amount of food reserves and the type of food being collected (Weeks et al.

2004). This type of supercolony formation, however, may be exclusively limited to S. invicta.

In other ant species, the situation is much less clear and a number of hypotheses, mainly

derived from studies on the Argentine Ant Linepithema humile, are currently being discussed.

One hypothesis is that supercolonies could be formed by the fusion of smaller colonies, as

was observed in the strictly subterranean, non-invasive ant species Lasius austriacus (Steiner

et al. 2007). Colony fusion has also been observed in Linepithema humile colonies under

laboratory conditions (Vasquez and Silverman 2008; Vasquez et al. 2009) but empirical data

supporting this scenario in wild supercolonies is lacking. Theoretically, fusion of small

colonies to large supercolonies might be facilitated by two mechanisms, i.e. genetic similarity

(and thus similarity in heritable recognition cues) of introduced colony subsets (proposed by

Tsutsui et al. 2000, among others) or similarity of recognition cues irrespective of genetic

similarity, e.g. as a result of frequency-dependent selection against colonies bearing rare

recognition alleles (proposed by Giraud et al. 2002). In particular, Tsutsui et al. (2000)

suggested that supercolonies in Linepithema humile may likely be the consequence of severe

genetic bottlenecks accompanying the introduction and establishment of subsets of native

populations (termed ‘propagules’) into previously uninhabited territory (supported by data

from Brandt et al. 2009b). The reduced overall genetic variability, may likely be reflected by

limited phenotypic variability in the heritable cues involved in nestmate recognition (Adams

1991; Pennanec'h et al. 1997; Takahashi et al. 2001; Brandt et al. 2009b; van Zweden et al.

2009) which are often cuticular hydrocarbons (CHC’s)(Vander Meer and Morel 1998; Lahav

et al. 1999; Akino et al. 2004; Torres et al. 2007; Brandt et al. 2009a). The reduced variability

in CHC profiles, in turn, might thus facilitate the fusion of smaller, genetically and chemically

similar propagules into large supercolonies (Tsutsui et al. 2000; Payne et al. 2004; Vasquez

and Silverman 2008; Vasquez et al. 2009). However, since they detected low levels of

relatedness in the very same supercolonies that Tsutsui et al. (2000) studied, Giraud et al.

(2002) suggested that L. humile supercolonies might not have experienced a genetic


16

bottleneck. Alternatively, Giraud et al. (2002) suggested that a reduced genetic variability at

alleles involved in the biosynthesis of CHC’s (a potential prerequisite for tolerance in

supercolonies) may have been the result of a ‘genetic cleansing’ process in which frequency-

dependent selection against less common alleles after introduction might lead to genetically

heterogeneous but chemically homogenous supercolonies. Both scenarios, however, seem to

be incorrect. If L. humile supercolonies resulted from the fusion of unrelated small colonies,

large supercolonies should be genetically more diverse than small supercolonies, which is not

the case (e.g. Jaquiery et al. 2005). Furthermore, Pedersen et al. (2006) were able to show that

in its native range, L. humile also forms supercolonies comprising unrelated individuals,

implying that the formation of supercolonies in L. humile may not a phenomenon that follows

upon introduction into a new habitat.

Thus, the emergence of large L. humile supercolonies in its introduced range may be the result

of a developmental transformation rather than an evolutionary one (Helanterä et al. 2009).

Assuming that colony-fusion is an inappropriate hypothesis, supercolonies inevitably start

small, i.e. from small colony fragments (propagules) or single mated queens dispersing on the

wing. As a consequence, genetic bottlenecks are likely to be a common starting-point among

supercolonies (Helanterä et al. 2009). Like in other biotic invasions, relaxed ecological

constraints such as release from predators or pathogens may facilitate supercolony growth

(Holway et al. 2002; Cremer et al. 2008) until it is hampered by rivalling supercolonies or

unsuitable habitat. The mode by which supercolonies usually grow (i.e. intranidal mating and

colony budding) is a mix of reproduction and dispersal strategies that should favour family-

based supercolonies, as is supported by high relatedness values within supercolonies of

various invasive and non-invasive ant species (Helanterä et al. 2009).

Focal species and aim of the study

Anoplolepis gracilipes (Smith)(Formicidae, Formicinae), formerly known as Anoplolepis

longipes (Emery 1925), Formica longipes (Jerdon 1851) and Plagiolepis longipes (Emery

1887), is an invasive ant species that occurs over large areas in South-East Asia and the

Indopacific islands and archipelagos (reviewed in Wetterer 2005). This ant species has

received worldwide attention due to its local displacement of a key-stone omnivore, the

endemic red land crab Gecarcoidea natalis on Christmas Island, Indian Ocean (O'Dowd et al.

2003), leading to structural changes in the island forest ecosystem (Davis et al. 2008; Davis et

al. 2010). Apart from its negative impact on individual invertebrate species (McNatty et al.


17

2009) and even entire invertebrate communities (Holway et al. 2002; Hill et al. 2003; Gerlach

2004; Lester and Tavite 2004), infestations of A. gracilipes may also affect vertebrate species

such as reptiles (Feare 1999) and birds (Gerlach 2004; Davis et al. 2008; Matsui et al. 2009;

Davis et al. 2010; Lach and Hooper-Bùi 2010).

Despite detailed information on its current distribution, its origin remains elusive (Wetterer

2005). While some authors suggested that A. gracilipes may be native to some parts of

tropical Asia, especially India (due to the oldest records), others considered East Africa as the

origin of this species as the genus Anoplolepis is almost exclusively African (e.g. Bolton

1995; reviewed by Wetterer 2005). Historical records, ecological data and the repeated

detection of new incursions demonstrate that A. gracilipes is introduced and invasive in Japan

(Matsui et al. 2009), Australia (Hoffmann and Saul 2010), New Zealand (Ward et al. 2006)

and the Indopacific region, e.g. on Hawaii (Krushelnycky et al. 2005; Kirschenbaum and

Grace 2007), the Seychelles (Gerlach 2004), Tokelau (Lester and Tavite 2004; Abbott et al.

2007) and Christmas Island (O'Dowd et al. 2003; Thomas et al. 2010a). In South-East Asia, A.

gracilipes thrives especially in disturbed habitats such as private orchards and corporate oil

palm plantations on Borneo (Pfeiffer et al. 2008; Brühl and Eltz 2009; Drescher et al. 2011),

cacao agroforests on Sulawesi (Bos et al. 2008; Wanger et al. in press) and secondary forests

on Papua New Guinea (Klimes et al. 2011).

Like other invasive ant species such as Lasius neglectus (Ugelvig et al. 2008), Linepithema

humile (Jaquiery et al. 2005), Pheidole megacephala (Fournier et al. 2009), Solenopsis invicta

(Vander Meer and Porter 2001) and Wasmannia auropunctata (Le Breton et al. 2004), A.

gracilipes forms large, polygynous (multi-queens) and polydomous (multiple nests)

supercolonies of extremely high worker densities (Abbott 2005; Abbott 2006; Lester et al.

2009). Queens usually mate within the nest (intranidal mating) and small-scale dispersal

occurs via budding, i.e. occupation of a suitable nest at the supercolony fringe by one or

several queens accompanied by workers and sometimes brood (‘propagule’) from the queens’

maternal nest (Holway et al. 2002). Despite the capture of winged males and females in

malaise traps on Christmas Island (Abbott 2006), active long-range dispersal by A. gracilipes

has yet to be observed. In contrast, passive long-range dispersal via private and commercial

human traffic (termed ‘human-mediated jump-dispersal’) is well documented (i.e. Harris et al.

2005) and may potentially represent A. gracilipes’ only form of dispersal over long distances.


18

Despite the numerous scientific publications describing its ecological impact in the

Indopacific region, A. gracilipes belongs to the lesser studied invasive ant species,

particularly with regard to its origin, its ecology in the native range and the way by which

supercolonies are structured. With this thesis I aim to fill some of these gaps in order to

contribute to the general insights on the mechanisms of biological invasions and its

consequences for biodiversity and community composition. In particular, I studied genetic

relationships between populations in one of A. gracilipes’ putative native ranges (South-East

Asia) and its introduced range (Indopacific region) to shed light on A. gracilipes’ origin. If A.

gracilipes’ origin were South-East Asia, its population there ought to be characterized by

considerably higher genetic diversity than the population in the Indopacific region, and

individual matrilines might be found in both regions, indicating rather recent transfer. Based

on the observations from other invasive ant species (see above), however, I did not expect to

find a difference in social structure between both regions. I furthermore studied A. gracilipes’

role in local ant communities as well as its interactions with ant guests in North-East Borneo

to learn more about its ecology in a putative native range. If South-East Asia was A.

gracilipes’ native region, I would expect to find comparably small supercolonies (as in

Linepithema humile’s native region, Pedersen et al. 2006) and potentially many mutualistic

and antagonistic interspecific relationships, such as species-specific trophobioses or

coevolved parasites. In turn, if A. gracilipes had been introduced to South-East Asia, I would

expect larger and ecologically destructive supercolonies similar to those in the Indopacific.

However, large populations of A. gracilipes and ecological damage alone would be too little

evidence to infer that A. gracilipes’ was introduced to South-East Asia, as native species

adapted to naturally disturbed ecosystems may become serious pests if anthropogenic land use

results in dramatically increased availability of suitable habitat. Thus, I measured behavioural,

genetic and chemical properties of different A. gracilipes supercolonies to gain insights into

its population structure in a potential native region. Furthermore, detailed information on the

structure of A. gracilipes may hold important information on the transferability of

mechanisms of supercolony-formation identified in a few focal species (above all L. humile

and S. invicta) to other invasive and non-invasive supercolonial ant species. And finally, as

supercolonialism in ants may be an evolutionary dead end (Helanterä et al. 2009), revealing

the structure of supercolonies and the interactions between them may provide insights into

their evolutionary future.

IV. Synopsis

19

IV. THE ECOLOGY AND POPULATION STRUCTURE OF THE

INVASIVE YELLOW CRAZY ANT ANOPLOLEPIS GRACILIPES

– A SYNOPSIS

The global spread and establishment of invasive ant species poses a serious thread to

ecosystems worldwide. While numerous reports describe local ecological impacts of many

invasive ant species, most of our knowledge on the mechanisms behind ant invasions is

derived from research on two focal species, the Argentine Ant Linepithema humile and the

Red Imported Fire Ant Solenopsis invicta (Holway et al. 2002). However, in order to

understand biological invasions in general and invasive ants in particular, studies on other

invasive ant species are required.

The primary objective of this study was to contribute to the insights on biological invasions

by studying the invasive Yellow Crazy Ant Anoplolepis gracilipes. In particular, I explored

the genetic relationship between A. gracilipes populations from its introduced range

(Indopacific) and one of its putative native ranges (South-East Asia), providing first genetic

evidence of A. gracilipes’ potential introduction history I further examined behavioural,

genetic and chemical properties of different A. gracilipes supercolonies, thus revealing A.

gracilipes’ mode of supercolony formation as well as the evolutionary potential of

supercolonialism in an invasive ant. And finally, I studied the ecology of A. gracilipes in

South-East Asia, in particular its interactions with the local ant fauna and its association with

several ant guest species, and discuss my findings with regard to the study area as a potential

native region of this species.

In CHAPTER I, I present data on the genetic diversity of Anoplolepis gracilipes populations

in South-East Asia and the Indopacific region. The current distribution of A. gracilipes

stretches across tropical regions worldwide, with the majority of records stemming from

South-East Asia and the Indopacific (Fig. 1). While there is no doubt that A. gracilipes has

been introduced to large parts of the Indopacific, its native region in general and its status in

South-East Asia in particular remain unclear (Wetterer 2005). Mitochondrial DNA sequences

(mtDNA ‘haplotypes’) of workers from 118 locations in 15 countries revealed considerably

high haplotype diversity but very low nucleotide diversity in both the South-East Asian and

the Indopacific population, suggesting that A. gracilipes may have experienced severe genetic

bottlenecks in both regions followed by rapid population growth. Despite being low in both

IV. Synopsis

20

Figure 1. Global distribution of the invasive Yellow Crazy Ant Anoplolepis gracilipes. The colour coding represents Anoplolepis gracilipes’ invasional status in tropical regions worldwide with South-East Asia indicated as one of the potential native ranges (with kind permission from Darren Ward, Landcare Research, New Zealand, 2005).

regions, nucleotide diversity in the Indopacific was half the nucleotide diversity of the South-

East Asian population, implying that the genetic bottleneck experienced by the Indopacific

population may have been more severe and may have occurred more recently. Taking into

account the detection of matching haplotypes between the South-East Asian and the

Indopacific population, I suggest that A. gracilipes may have been first introduced to South-

East Asia from a yet unidentified native region (possibly from India in the early 19th century)

and subsequently to the Indopacific islands from South-East Asian source populations.

In CHAPTERS II and III, I explore the population structure of Anoplolepis gracilipes as well

as its evolutionary implications based on behavioural, genetic and chemical analyses of

several supercolonies. CHAPTER II contains a detailed investigation of six A. gracilipes

supercolonies from Poring Hot Springs in North-East Borneo, combining data from

behavioural bioassays, analysis of nuclear microsatellite markers and GC-MS analyses of

cuticular hydrocarbon (CHC) profiles (Drescher et al. 2010). The study indicated that A.

gracilipes supercolonies contain very closely related individuals, with relatedness values

between workers even exceeding those that would be expected in monogynous colonies of

singly mated queens. Supercolonies of A. gracilipes thus differ substantially from

supercolonies of Linepithema humile and Solenopsis invicta, which may contain unrelated

individuals (Bernasconi and Strassmann 1999; Pedersen et al. 2006). The extreme genetic

similarity between individuals in A. gracilipes supercolonies is likely due to almost exclusive

intranidal mating (i.e. inbreeding in the case of A. gracilipes) and an unusual reproductive

IV. Synopsis

21

Figure 2. Intercolonial aggression between workers of Anoplolepis gracilipes. In an initial phase of aggression between A. gracilipes workers from different super-colonies, workers will often grasp the intruder by its legs and antennae. (Photo: J. Drescher)

system that may involve clonal reproduction of queens (Drescher et al. 2007). Two pairs of

the six spatially separated supercolonies were closely related, showed high cuticular

hydrocarbon (CHC) profile similarity and tolerated each other in bioassays, suggesting that

they represented established, anthropogenically dispersed fragments of the same supercolony.

With the exception of these two pairs of mutually tolerant supercolony fragments, workers

from different supercolonies were highly aggressive towards each other (Fig. 2) and were

profoundly differentiated both genetically and

chemically. In particular, several supercolonies

contained private microsatellite alleles, and CHC

profiles of mutually aggressive supercolonies showed

qualitative differences, i.e. the profiles partially

comprised different compounds. The extent of

qualitative differences between CHC profiles of

different A. gracilipes supercolonies is remarkable,

because in most ants, CHC profiles differ quantitatively

between conspecific colonies and qualitatively between

species (Vander Meer and Morel 1998). Overall, the

data suggested that gene flow between mutually

aggressive A. gracilipes supercolonies may be extremely

limited or even absent. A positive feedback between

genetic, chemical and behavioural traits may further

enhance supercolony differentiation through genetic drift

and neutral evolution.

CHAPTER III contains a separate study which confirmed the results and implications of the

aforementioned study (CHAPTER II). The study was conducted in a different NE-Bornean

region (Sepilok Forest Reserve) and comprised behavioural, genetic and chemical analyses of

eleven local supercolonies as well as three supercolonies from Poring Hot Springs. In addition

to the analysis of nuclear microsatellites (CHAPTER II), I examined mtDNA sequences of the

14 supercolonies (see CHAPTER I) to study the extent of genetic differentiation between

supercolonies. The study revealed that intercolonial differentiation in mitochondrial DNA

corresponded to microsatellite and CHC profile differentiation as well as patterns of

aggression between different supercolonies. Gene flow between different A. gracilipes

supercolonies is extremely limited or even absent (Drescher et al. 2010; Thomas et al. 2010a),

IV. Synopsis

22

Figure 3. Aggression of Ano-

plolepis gracilipes workers towards sexuals from different supercolonies. (A) Worker aggression towards allocolonial males and (B) worker aggression towards allocolonial queens (Photos: J. Drescher)

which may be largely due to strict intranidal mating and absence of nuptial mating flights

(Drescher et al. 2007; Drescher et al. 2010). Hence, the emigration of males and queens from

their own supercolony to foreign supercolonies may represent the only way by which gene

flow between different supercolonies may be maintained. Behavioural assays between

workers and sexuals (males/queens) from mutually aggressive supercolonies, however,

revealed pronounced aggression of workers towards males and queens from foreign

supercolonies (Fig. 3A, 3B), thus possibly impairing gene flow between different

supercolonies by means of inhibiting the migration of sexuals between supercolonies. As gene

flow between different supercolonies may have been reduced or even been absent for quite

some time, A. gracilipes supercolonies may already be genetically and chemically sufficiently

different to hamper mating and colony founding by males and queens from different

supercolonies. In a series of preliminary breeding experiments using laboratory colonies, I

tested whether queens still retained the capability to mate with males from different

supercolonies and whether queens fertilized in such a way were able to start new colonies.

The experiments demonstrated that despite mating between allocolonial sexuals occurred (as

confirmed by filled spermathecae of queens that were raised as virgins) and despite the fact

that the queens laid eggs, no viable worker offspring was produced within six months, which

eventually led to the demise of the experimental subcolonies. Thus, reproductive isolation

between different A. gracilipes supercolonies may possibly already impede gene flow

between them, which should consequently lead to the evolutionary divergence of different A.

gracilipes supercolonies into different species.

In CHAPTERS IV and V, I focus on A. gracilipes’ ecology. CHAPTER IV describes the

ecological dominance of Anoplolepis gracilipes within an ant community in North-East

Borneo (Drescher et al. 2011). Like most invasive organisms, A. gracilipes is generalistic and

opportunistic in both feeding and nest site preferences. This ant is omnivorous and can

IV. Synopsis

23

Figure 4. Feeding preferences and interspecific aggression of Anoplolepis gracilipes. As a generalist feeder, A. gracilipes gathers nutrients from various sources such as (A) sugar-rich honeydew from hemipteran trophobionts and (B) protein from prey items or carrion. A. gracilipes is highly aggressive towards other ants such as (C) Dolichoderus thoracicus, and dead ants from other species can frequently be found on refuse piles inside A. gracilipes nests. (Photos by J. Drescher)

frequently be observed foraging honeydew (Fig. 4A) and both live and dead arthropods (Fig.

4B). Mimicking sugar- and protein-based food resources with honey and tuna baits, I could

demonstrate that within supercolonies, ant diversity was much lower than outside

supercolonies. Moreover, A. gracilipes detected food resources quicker than other ants and

aggressively defended them until they were fully exploited. This combination of early

detection and subsequent monopolization of food items has also been described in the

invasive ant Linepithema humile (Davidson 1998; Holway 1999) and may trigger the release

from the dominance-discovery trade-off, which describes that ants are subject to a trade-off

between rapid resource discovery and subsequent domination of resource exploitation. The

dominance-discovery trade-off is perceived as a mechanism facilitating species coexistence,

thus breaking the trade-off may promote A. gracilipes’ ecological dominance in ant

communities and the potential displacement of individual ant species within its supercolony

boundaries. However, the worker density in an ant colony and its competitive abilities in

resource discovery and monopolization are linked (Oliver et al. 2008), thus making it difficult

to disentangle whether breaking the trade-off is a cause or rather a consequence of high

worker densities in A. gracilipes supercolonies. The study further suggested that although A.

gracilipes’ potential capability to displace native ants and its superiority in resource

acquisition may be intertwined, its ecological success in NE-Borneo may be fundamentally

based to its elevated interspecific aggression. In arena bioassays, A. gracilipes readily

engaged in mortal encounters with seven native ant species (including Dolichoderus

thoracicus, Fig. 4C) and survived significantly longer than the majority of the species tested,

occasionally even when being outnumbered tenfold (e.g. in bioassays against Crematogaster

coriaria). The weaver ant Oecophylla smaragdina was the only ant species being able to

IV. Synopsis

24

Figure 5. Myrmecophilus

pallidithorax, one of the social parasites of Anoplolepis gracilipes. (A) Intact cricket from top view and (B) manipu-lated cricket being attacked by A. gracilipes workers. (Photos: J. Drescher)

withstand A. gracilipes’ elevated interspecific aggression, potentially due to its size and its

increased aggressive behaviour as a dominant ant species. However, other large, feisty ant

species such as Diacamma sp. or Odontomachus sp. were frequently attacked and pursued by

individual A. gracilipes workers, overall suggesting that A. gracilipes may dispose of superior

fighting skills compared to many ant species of comparable size.

CHAPTER V deals with the association between Anoplolepis gracilipes and a cleptoparasitic

ant guest, the myrmecophilous cricket Myrmecophilus pallidithorax (Fig. 5A, 5B). Among

several ant guests (such as the Handsome Fungus Beetle Trochoideus desjardinsi, two

unidentified species of tropical log beetles of the genus Aphanocephalus and two species of

gnat bugs, Oncylocotis spp., see Štys et al. 2010), Myrmecophilus pallidithorax was found in

the majority of A. gracilipes supercolonies in NE-Borneo that were screened for ant guests.

Behavioural experiments involving the exchange of intact and manipulated crickets

(manipulated = hind legs removed by induced autotomy to reduce mobility) between

experimental subcolonies from different A. gracilipes supercolonies indicated that A.

gracilipes was capable of detecting the cricket, but failed to catch and kill it due to the

supreme agility of the cricket. The experiments further suggested that chemical deception may

play an important role in the infestation of A. gracilipes supercolonies by M. pallidithorax, as

the mortality of manipulated crickets was slightly lower when ants and crickets were sampled

from the same supercolony than when sampled from different supercolonies. Finally,

preliminary data of cuticular hydrocarbon (CHC) profiles revealed considerable congruency

between CHC profiles of A. gracilipes and M. pallidithorax. The lack of replication across

several A. gracilipes supercolonies, however, does to allow conclusions as to whether M.

pallidithorax uses chemical mimicry or chemical camouflage to reduce aggression from its

host.

VI. mtDNA diversity of Anoplolepis gracilipes

25

V. MITOCHONDRIAL DNA DIVERSITY OF THE YELLOW CRAZY

ANT ANOPLOLEPIS GRACILIPES IN SOUTHEAST ASIA AND

THE INDOPACIFIC

ABSTRACT

Identifying the native ranges and invasional pathways of invasive organisms is crucial for the

development of management strategies and preventive measures. This study focused on the

invasive Yellow Crazy Ant Anoplolepis gracilipes, of which the native range is unclear. We

compared samples from the introduced range (Indopacific region) to samples from one of its

putative native ranges (South-East Asia) to resolve whether A. gracilipes might be native to

SE-Asia and to identify the origin of the Indopacific colonies. We analyzed mtDNA

sequences of 118 individuals obtained from different nests. After genotyping with two

mitochondrial DNA markers, we calculated genetic diversity using standard indices and

constructed a haplotype network. Haplotype diversities in the SE-Asian and the Indopacific

samples were similarly high. Nucleotide diversity, however, was low SE-Asia and even lower

in the Indopacific, indicating recent genetic bottlenecks followed by rapid population growth

in both regions. Overall, we detected 29 haplotypes, of which three were shared by samples

from both SE-Asia and the Indopacific. Both regions contained large proportions of private

haplotypes as well as several singletons (haplotypes found in only one sample). Our data

indicate that both the SE-Asian and the Indopacific population of A. gracilipes may have

experienced a genetic bottleneck, which may have been more severe in the Indopacific. This

suggests that A. gracilipes may have initially been introduced to SE-Asia followed by

subsequent introduction to the Indopacific from SE-Asian populations, which is supported by

matching haplotypes in samples from both regions. Furthermore, A. gracilipes populations in

both regions may be subject to diversification processes, leading to the evolution of novel

mtDNA lineages. We suggest sampling and genotyping A. gracilipes samples from other

putative native ranges (e.g. southern India) to confirm these findings.


26

INTRODUCTION

Biological invasions threaten local biodiversity in many habitats worldwide through

negatively affecting native taxa (e.g. Kenis et al. 2009), thus frequently entailing both

economic and environmental damage in the regions affected (Pimentel et al. 2005; Ehrenfeld

2010; Perrings et al. 2010). In order to reduce the impact of biotic invasions, studying

invading species within their home ranges is crucial to develop biocontrol strategies through

identifying potential antagonistic organisms (Messing and Wright 2006). Likewise, locating

the origin of an invasive organism is imperative for devising preventive measures such as

intercepting invasional pathways by targeting vectors of invasion (Mack et al. 2000a; Myers

et al. 2000; Carlton and Ruiz 2005; Geller et al. 2010). Hence, knowing where an invader

comes from, which community it evolved in and how it was introduced into the invaded

habitat may enable us not only to understand the mechanism of its invasion, but further

facilitates the development of both management strategies and preventive measures. Yet there

are numerous invasive species whose native range or introduction pathway is unknown

(Geller et al. 2010). Similarly, there is a vast number of cryptogenic species (sensu Carlton

1996), i.e. species that cannot be labeled as being either native or introduced in a given habitat

due to lack of sufficient biogeographic, historical, systematic or ecologic data (Hewitt et al.

2004; Neves and da Rocha 2008; Geller et al. 2010; Ignacio et al. 2010). The recent progress

in molecular markers, however, has provided adequate genetic tools to target these questions

in invasion biology, as has repeatedly been shown by retrospectively constructing individual

invasion histories and introduction pathways of invasive organisms (Jousson et al. 2000;

Durka et al. 2005; Grapputo et al. 2005; Corin et al. 2007; Caldera et al. 2008; Ugelvig et al.

2008; Iriarte et al. 2009; Foucaud et al. 2010).

In this study we analyzed the genetic relationship between supercolonies of the invasive

Yellow Crazy Ant Anoplolepis gracilipes from Southeast Asia and the Indopacific region.

This invasive ant species has been brought to worldwide attention due to its local

displacement of a key-stone omnivore, the endemic red land crab Gecarcoidea natalis on

Christmas Island, Indian Ocean (O'Dowd et al. 2003), leading to structural changes in the

island forest ecosystem (Davis et al. 2008; Davis et al. 2010). Apart from its negative impact

on individual invertebrate species (McNatty et al. 2009) and even entire invertebrate

communities (Holway et al. 2002; Hill et al. 2003; Gerlach 2004; Lester and Tavite 2004),

infestations of A. gracilipes may also affect vertebrate species such as reptiles (Feare 1999)


27

and birds (Gerlach 2004; Davis et al. 2008; Matsui et al. 2009; Davis et al. 2010; reviewed in

Lach and Hooper-Bùi 2010). Like other invasive ant species such as Lasius neglectus

(Ugelvig et al. 2008), Linepithema humile (Jaquiery et al. 2005), Pheidole megacephala

(Fournier et al. 2009) and Wasmannia auropunctata (Le Breton et al. 2004), A. gracilipes

forms large, polygynous (multi-queens) and polydomous (multiple nests) supercolonies of

extremely high worker densities (Abbott 2005; Abbott 2006; Lester et al. 2009). Queens

usually mate within the nest (intranidal mating) and small-scale dispersal occurs via budding,

i.e. occupation of a suitable nest at the supercolony fringe by one or several queens

accompanied by workers and sometimes brood (termed propagule) from the queens’

aboriginal nest (Holway et al. 2002; Drescher et al. 2010). Active long-range dispersal is

unknown in this species. However, passive long-range dispersal occurs nonetheless as

propagules of this species are inadvertently introduced into new habitats via human-mediated

dispersal, e.g. in soil, agricultural products, timber, packaging material, sea containers or

automobiles (both local traffic and shipments)(Harris et al. 2005, pers. obs.). The current

distribution of A. gracilipes is almost exclusively limited to tropical regions worldwide, with

the highest density of populations found in Southeast Asia and the Indopacific region

(Wetterer 2005). However, the origin of this species, which was first described as Formica

longipes in India (Jerdon 1851), remains unclear. Several authors suggested that A. gracilipes

may be native to some parts of tropical Asia, especially India (due to the oldest records),

while others considered East Africa as the origin of this species as the genus Anoplolepis is

known to be almost exclusively African (see Wetterer 2005 for references). Historical

records, ecological data and the repeated detection of new incursions demonstrate that A.

gracilipes is introduced and invasive in Japan (Matsui et al. 2009), Australia (Hoffmann and

Saul 2010), New Zealand (Ward et al. 2006) and the Indopacific region, e.g. on Hawaii

(Krushelnycky et al. 2005; Kirschenbaum and Grace 2007), the Seychelles (Gerlach 2004),

Tokelau (Lester and Tavite 2004; Abbott et al. 2007) and Christmas Island (O'Dowd et al.

2003; Thomas et al. 2010a).

In all regions including the potential native range of SE-Asia, A. gracilipes thrives especially

in disturbed habitats. High-density supercolonies have been reported from orchards and oil

palm plantations on Borneo (Pfeiffer et al. 2008; Brühl and Eltz 2009; Drescher et al. 2011),

cacao agroforests on Sulawesi (Bos et al. 2008; Wanger et al. in press) and secondary forests

on Papua New Guinea (Klimes et al. 2011). This almost exclusive limitation of its distribution

to anthropogenically disturbed habitats and its ecological dominance may indicate that A.


28

gracilipes is invasive in SE-Asia. Alternatively, A. gracilipes may potentially be a native

species adapted to naturally disturbed habitats (e.g. annually flooded swamp forests along the

river Kinabatangan in Borneo (Drescher et al. 2007), which has become increasingly

abundant due to an increasing availability of suitable, anthropogenically disturbed habitat.

Thus, in this study, we investigated whether A. gracilipes may have originated in Southeast

Asia by analyzing mitochondrial DNA diversity in samples from both the introduced and the

putative native range. In particular, based on the repeated observation that invasive

populations experience severe genetic bottlenecks upon introduction (e.g. Tsutsui et al. 2000;

Suarez et al. 2008; Geller et al. 2010), we should expect to detect higher genetic variability in

the native range (potentially SE-Asia) than in the introduced range (Indopacific). Taking into

account that the introduction of A. gracilipes to some Indopacific habitats has occurred within

the past 100 years (O'Dowd et al. 1999; O'Dowd et al. 2003) and still occurs up to date

(Harris et al. 2005), we should furthermore expect to find matching haplotypes between SE-

Asia and the Indopacific, thus enabling us to identify pathways of introduction. Finally,

provided there are populations that have been established long enough to acquire novel

mutations in situ (as may be the case on Christmas Island, Thomas et al. 2010a), we should

expect to find unique haplotypes in the introduced range that cannot be found in the putative

native range.

METHODS

Anoplolepis gracilipes workers were sampled from a total of 118 locations in 15 countries

throughout Southeast Asia and the Indopacific region (Fig. 6). Malaysian Borneo is

overrepresented in this data set, as detailed studies focusing on the population structure,

behaviour and ecology of A. gracilipes were conducted in this area (Drescher et al. 2007;

Drescher et al. 2010; Stys et al. 2010; Drescher et al. 2011). Individuals were collected near

the nest entrance to ensure supercolony affiliation and immediately transferred into 99.8%

EtOH p.A..

As preliminary surveys revealed that there was never more than one mtDNA haplotype to be

found per supercolony, we extracted DNA from a single worker per location (putative

supercolony) using the PUREGENE® DNA purification kit. The obtained DNA pellet was

resuspended in 50µl double distilled water and stored at -20°C. Mitochondrial DNA was


29

Figure 6. Locations of Anoplolepis gracilipes sampling sites throughout Southeast Asia and the Indopacific region. Sampling site labels are arranged according to the invasional status of A. gracilipes in its known range (Wetterer 2005), i.e. invasive in the Indopacific (lower right) and potentially native in Southeast Asia (upper left).

amplified using the primer pairs Horst (5′-AC(TC)-ATACTTTTAACTGATCG-3′) designed

by D. Kronauer (unpublished)/Ben (5′-C(AT)AC(AT)AC(AG)TAATA(GT)GTATCATG-3′)

(Moreau et al. 2006) for partial Cytochromeoxidase I (COI) (Genbank Accession nos

DQ888821–DQ888825) corresponding to positions 2407–2427 (Horst) and 2891–2914

(BEN) relative to the mitochondrial genome of Apis mellifera (Crozier and Crozier 1993) and

CBI/CBII for partial cytochrome b (cytb) (Genbank Accession nos DQ888817–DQ888820)

(Crozier et al. 1991).

PCR was performed in a total reaction volume of 25 µl containing approximately 10 ng of

template DNA, 1x PCR buffer, 2 mM MgCl2, 240 µM dNTP’s, 800 µM of both forward and

reverse primer and 1.2 U of Taq DNA polymerase (MolTaq byMolzym GmbH or Genaxxon

BioScience Taq-Polymerase). Amplification via PCR of both fragments of mtDNA was

performed either in an Eppendorf or Biometra thermocycler at the following conditions: 3

min at 94°C, followed by 30 cycles of 94°C for 1 min, 1 min for the annealing step at 45°C,

1.5 min at 72°C and a final extension of 3 min at 72°C.


30

The purified mtDNA fragments were sequenced by SEQLAB sequence laboratories

(Göttingen, Germany), analyzed and aligned using BioEdit version 7.0.5.3 (Hall 1999).

Genetic diversity within each region (SE-Asia and Indopacific) was estimated by calculating

haplotype diversity h (probability that two randomly chosen haplotypes are different) and

nucleotide diversity π (mean proportion of nucleotides differing in all pairwise combinations)

using DnaSP ver. 5.10.01 (Librado and Rozas 2009). High h and high π values indicate large,

stable populations with long evolutionary histories or secondary contact between population

subsets, high h and low π suggest a recent bottleneck followed by rapid growth and

accumulation of mutations and low h and low π suggest a recent bottleneck or founder event

by a single or few lineages (Grant and Bowen 1998; Quek et al. 2007). Additionally, a

haplotype network was constructed based on the concatenated sequence of 904 bp of

mitochondrial DNA (consisting of a 460 bp fragment of the COI gene and a 444 bp fragment

of cytb) using the 95% parsimony algorithm implemented in TCS 1.21 (Clement et al.

2000b).

RESULTS

We detected 29 mtDNA haplotypes in the 118 Anoplolepis gracilipes samples from SE-Asia

and the Indopacific (Fig. 7). Sixteen of the 29 haplotypes were singletons, i.e. were found in

just a single sample, while the other 13 haplotypes were shared by two to 26 samples. The

SE-Asian population contained 11 of these singletons, while only five singletons were found

Sampling region N Nht Nph Nps h SDh π (%)

SE-Asia 56 18 15 28 0.847 ±0.033 0.69

Indopacific 62 14 11 15 0.867 ±0.018 0.37

Total 118 29 - 30 0.966 ±0.013 0.66

Table 1. Genetic diversity measures of Anoplolepis gracilipes based on mtDNA haplotypes of samples from Southeast Asia (putative native range) and the Indopacific region (introduced range). N, number of samples per region; Nht, number of haplotypes per region; Nph, number of private haplotypes per region; Nps, number of polymorphic sites; h, haplotype diversity; SDh, standard deviation; π, nucleotide diversity (in %).


31

Figure 7. 95% parsimony haplotype network of Anoplolepis gracilipes mtDNA sequences from SE-Asia and the Indopacific region. The concatenated mtDNA sequence comprised 904 bp (460 bp of COI, 444bp of cytb) and was constructed using TCS 1.21 (Clement et al. 2000b). Samples within the same box share a common haplotype. The number of internodes separating different boxes indicates the number of mutations needed to transform one haplotype into the other. Country abbreviations are given according to ISO standard 3166-1. The size of the circles is equivalent to the number of samples sharing a haplotype.

in the Indopacific population. Overall genetic diversity was high (h = 9.966 ± 0.013, π = 0.66

%, Tab. 1). Samples from the SE-Asian population contained more private haplotypes (15 in

SE-Asia vs. 11 in the Indopacific) and polymorphic sites (28 in SE-Asia vs. 15 in the

Indopacific) as the Indopacific population and showed almost twice the nuclear diversity (π SE-

Asia = 0.69 % vs. π Indopacific = 0.37 %). Haplotype diversity, however, was similarly high in

both the Indopacific and the SE-Asian population (Tab. 1). Three of the 29 haplotypes were

shared by samples from both the Indopacific and the SE-Asian population. In detail, these

three haplotypes contained samples from (1) Malaysia/Brunei/Philippines and

Tokelau/Fiji/Tuvalu/Australia, (2) Malaysia and Hawaii, USA and (3) Thailand and Australia

(Fig. 7, 8).


32

Figure 8. Anoplolepis

gracilipes mtDNA haplotypes in relation to their geographical origin. Samples are color-coded as stemming from either A. gracilipes’ introduced range (Indopacific region, red) or its putative native range (Southeast Asia, green) according to Wetterer (2005). Country abbreviations are given according to ISO 3166-1. The size of the circles is equivalent to the number of samples sharing a haplotype.

DISCUSSION

This study explored the genetic relationship between Anoplolepis gracilipes populations from

its introduced range in the Indopacific and one of its putative native ranges, Southeast Asia,

using two mitochondrial DNA markers. The genetic analyses could not unambiguously

identify SE-Asia as the native range of A. gracilipes, but strongly suggested that many of the

A. gracilipes supercolonies found in the Indopacific today may have been introduced from

there. Sampling from other potential native ranges of A. gracilipes, especially southern India

(Wetterer 2005) is thus crucial in revealing the origin of this ant, as well as clarifying its

status (invasive vs. native) in SE-Asia.

Previous research revealed that A. gracilipes supercolonies never contain more than one

haplotype (unpub. results, Abbott et al. 2007; Thomas et al. 2010a) and supercolonies bearing

the same haplotype may often belong to different supercolonies (Drescher et al. 2007;

Drescher et al. 2010). Thus, our finding of 29 haplotypes in 118 A. gracilipes samples

represents the minimum number of supercolonies contained in the data set.

Despite the geographic bias in our data set and unequal sample sizes for both geographic

subgroups, the haplotype diversity was similarly high in both the SE-Asian and the

Indopacific range of A. gracilipes and was only slightly lower than the haplotype diversity in

a worldwide sampling of the Argentine Ant Linepithema humile (h = 0.92, (Corin et al.


33

2007)). In turn, overall nucleotide diversity π was very low. In concert with high haplotype

diversity h, low π values indicate a severely bottlenecked population after a phase of rapid

growth (Grant and Bowen 1998; Quek et al. 2007). Moreover, the extremely low nucleotide

diversity of the Indopacific population indicates that the genetic bottleneck upon introduction

of A. gracilipes to the Indopacific may have been a lot more severe than upon the potential

introduction to SE-Asia. Alternatively, the severe difference in nucleotide diversity between

the SE-Asian and the Indopacific population may alternatively suggest a step-wise

introduction history (first bottleneck upon introduction to SE-Asia, second bottleneck upon

introduction to Indopacific from already bottlenecked SE-Asian sources).

Haplotype networks and phylogenetic approaches based on mtDNA have repeatedly been

used to identify introduction pathways of other invasive ant species such as Linepithema

humile (Corin et al. 2007), Solenopsis invicta (Caldera et al. 2008) or Wasmannia

auropunctata (Mikheyev and Mueller 2007). In these studies, introduction pathways were

identified by finding haplotypes or phylogenetic branches that were shared by samples from

the native range and the introduced range, respectively. In the present study, we found three

haplotypes that were shared by samples from A. gracilipes’ putative native range (SE-Asia)

and its introduced range (Indopacific)(Fig. 7, 8). This observation implies that the respective

supercolonies were introduced to Hawaii, Australia, Fiji, Tokelau and Tuvalu from the same

source supercolonies from A. gracilipes populations on Borneo, Thailand and the Philippines.

This scenario is supported by interception records at Australian borders which show that

roughly half of the A. gracilipes propagules that reached Australia between 1986 and 2003

were detected on cargo ships or planes coming from SE-Asia, while most other were from the

Indopacific (Harris et al. 2005). This, however, does not necessarily mean that A. gracilipes is

native to SE-Asia, as SE-Asia may have been colonized by A. gracilipes supercolonies from

yet another potential native range such as India or East Africa (Wetterer 2005), thus

constituting a transit region from where A. gracilipes is further introduced into habitats in the

Indopacific. As both India as well as several regions in SE-Asia (incl. Malaysia, Brunei,

Myanmar/Burma, Java) were connected by naval trade routes as parts of the British Empire

from as early as the 19th century on, introductions of A. gracilipes from India to SE-Asia

followed by local establishment of invasive populations in SE-Asia could have occurred

frequently over the past 150 to 200 years. As trade and commerce increased both within SE-

Asia as well as between SE-Asia and the Indopacific in the past decades, these potentially


34

invasive SE-Asian populations could now have become the source of many of the

introductions of A. gracilipes into Indopacific habitats.

Aside from haplotypes that were shared between samples from SE-Asia and the Indopacific,

we also found unique haplotypes in both regions, including haplotypes that were found in just

a single sample. Of these singletons, 11 were found in the SE-Asian population, while four

were found in the Indopacific. Large numbers of unique haplotypes in both SE-Asia and the

Indopacific may be due to incomplete sampling (i.e. the haplotypes have only been sampled

in one of the regions), differences in the propagules that have been introduced into both

regions from yet another unidentified native source population, or in situ mutations that differ

among regions (e.g. on Christmas Island, Thomas et al., 2010). The latter may be facilitated

by a combination of unusual mating and dispersal strategies. Intranidal mating seems to be the

exclusive reproductive strategy of A. gracilipes and gene flow between supercolonies is

largely absent (Drescher et al. 2010). As a result, A. gracilipes supercolonies may represent

isolated units following independent evolutionary trajectories. As novel mutations are likely

to spread more rapidly through small populations as opposed to large populations,

independent colony founding by dispersing alate queens (as observed by Abbott 2006) or

establishment of anthropogenically dispersed propagules may represent the two predominant

processes facilitating the emergence of supercolonies bearing novel mtDNA haplotypes.

Conclusion

This study provides first insights into the invasional biogeography of the invasive Yellow

Crazy Ant Anoplolepis gracilipes in SE-Asia and the Indopacific. Though failing in locating

A. gracilipes’ native region, our data imply that the SE-Asian population has experienced a

less severe bottleneck than the Indopacific population, suggesting that A. gracilipes may have

been introduced to the Indopacific from invasive populations in SE-Asia. Matching

haplotypes between SE-Asia and the Indopacific further suggest that SE-Asia may be origin

of many of the introduced supercolonies in the Indopacific. Finally, A. gracilipes populations

in both regions may be subject to diversification processes, potentially leading to novel

haplotypes that are maintained within supercolonies.


35

ACKNOWLEDGEMENTS

We thank Kirsti Abbott, Milan Janda, Robert Junker, Petr Klimes, Alain Lenoir, Juliane

Mooz, Thomas Schmitt, Deborah Schweinfest, Andrew V. Suarez and Darren Ward for their

contribution of A. gracilipes samples. This work was part of the Royal Society’s Southeast

Asia Rainforest Research Programme (SEARRP) and was funded by the Deutsche

Forschungsgemeinschaft DFG (SFB 554 ‘Evolution and mechanisms of arthropod behaviour’,

project E2).


36

VI. Genetic and chemical differentiation between supercolonies

37

VI. GENETIC AND CHEMICAL DIFFERENTIATION BETWEEN

ANOPLOLEPIS GRACILIPES SUPERCOLONIES

This chapter has been published as:

Drescher J, Blüthgen N, Schmitt T, Bühler J, Feldhaar H (2010) Societies drifting apart?

Behavioural, genetic and chemical differentiation between supercolonies in the Yellow Crazy

Ant Anoplolepis gracilipes. PLoS ONE 5(10): e13581

ABSTRACT

In populations of most social insects, gene flow is maintained through mating between

reproductive individuals from different colonies in periodic nuptial flights followed by

dispersal of the fertilized foundresses. Some ant species, however, form large polygynous

supercolonies, in which mating takes place within the maternal nest (intranidal mating) and

fertilized queens disperse within or along the boundary of the supercolony, leading to

supercolony growth (colony budding). As a consequence, gene flow is largely confined within

supercolonies. Over time, such supercolonies may diverge genetically and, thus, also in

recognition cues (cuticular hydrocarbons, CHC’s) by a combination of genetic drift and

accumulation of colony-specific, neutral mutations. We tested this hypothesis for six

supercolonies of the invasive ant Anoplolepis gracilipes in north-east Borneo. Within

supercolonies, workers from different nests tolerated each other, were closely related and

showed highly similar CHC profiles. Between supercolonies, aggression ranged from

tolerance to mortal encounters and was negatively correlated with relatedness and CHC

profile similarity. Supercolonies were genetically and chemically distinct, with mutually

aggressive supercolony pairs sharing only 33.1% ± 17.5% (mean±SD) of their alleles across

six microsatellite loci and 73.8% ± 11.6% of the compounds in their CHC profile. Moreover,

the proportion of alleles that differed between supercolony pairs was positively correlated to

the proportion of qualitatively different CHC compounds. These qualitatively differing CHC

compounds were found across various substance classes including alkanes, alkenes and

mono-, di- and trimethyl-branched alkanes. We conclude that positive feedback between

genetic, chemical and behavioural traits may further enhance supercolony differentiation


38

through genetic drift and neutral evolution, and may drive colonies towards different

evolutionary pathways, possibly including speciation.

INTRODUCTION

In most social insect species, a colony is a closed family unit that contains a queen and her

daughter workers (Hölldobler and Wilson 2009). The queen lays eggs, which are cared for

and reared by the workers. As the colony reaches maturity, males and virgin queens are

produced which mate with reproductives from other colonies in periodic mating events. Soon

after copulation, the males die while the fertilized queens disperse in an attempt to

independently found new colonies. As a consequence of this strategy, gene flow among

colonies is maintained within the population.

Social insects such as ants, however, show a large variety in social organization, reproductive

systems, mating behaviour and dispersal modes. Some tramp ant species such as Linepithema

humile (Jaquiery et al. 2005), Pheidole megacephala (Fournier et al. 2009), Wasmannia

auropunctata (Le Breton et al. 2004), Lasius neglectus (Ugelvig et al. 2008) or Anoplolepis

gracilipes (O'Dowd et al. 2003) share a similar set of strategies: They form large polydomous,

highly polygynous supercolonies, within which aggression is absent and individuals move

freely between nests (Holway et al. 2002). Supercolonies differ from other polygynous and

polydomous ant colonies, in that they are usually too large to allow direct cooperative

interactions of individuals from distant nests (Pedersen et al. 2006). In contrast to most other

ant species, nuptial flights of supercolony-forming ant species are rare or even absent, and

mating often takes place within the nest or the supercolony. After fertilization, queens stay

within the maternal nest, move to other nests within the supercolony or disperse at the fringe

of the supercolony through budding, i.e. the occupation of a suitable nest by one or several

queens that are accompanied by workers and sometimes brood from the colony the queens

themselves originated from. As a consequence of these mating and dispersal strategies, gene

flow between supercolonies may be extremely limited or absent, as has been pointed out by

studies on L. humile (Jaquiery et al. 2005; Pedersen et al. 2006; Thomas et al. 2006) and A.

gracilipes (Thomas et al. 2010b). The study by Thomas et al. (2010b) strongly suggests that

gene flow is absent between two sympatric and mutually aggressive A. gracilipes

supercolonies on Christmas Island, Indian Ocean. The two supercolonies were genetically


39

differentiated both in nuclear and mitochondrial loci. Furthermore, intranidal mating in A.

gracilipes can be frequently observed (in laboratory colonies of A. gracilipes) and workers are

highly aggressive towards virgin queens and males from other supercolonies (Drescher,

Feldhaar, pers. obs.), suggesting that gene flow among supercolonies may be prevented by

behavioural barriers. As a consequence, lack of gene flow between ant colonies should lead to

genetic differentiation among them.

Nestmate recognition in ants usually relies on cuticular hydrocarbons (CHC’s) (Vander Meer

and Morel 1998; Lahav et al. 1999; Akino et al. 2004; Torres et al. 2007; Brandstaetter et al.

2008) although other substance classes can also be involved, such as fatty acids (Franks et al.

1990). CHC’s are largely genetically determined (Adams 1991; Pennanec'h et al. 1997;

Takahashi et al. 2001; van Zweden et al. 2009; van Zweden et al. 2010), which implies that

genetic differentiation among sympatric ant colonies may entail the diversification of

genetically based CHC profile compounds through genetic drift and the accumulation of

supercolony-specific mutations. Thus, in ant species with strict intranidal mating, CHC-

profiles may differ between colonies not only in terms of relative abundances of epicuticular

compounds (quantitative differences, as is the case in most ant species with intercolonial

mating), but also in the composition of the profile itself (qualitative differences).

As the origin and future of supercolonialism in ants is unclear (Helanterä et al. 2009),

studying the degree and nature of differentiation within and between supercolonies may help

in understanding the evolutionary paths that those species might take (Helanterä et al. 2009).

We thus measured patterns of behavioural, genetic and chemical differentiation within and

between six spatially separated supercolonies in north-eastern Borneo, and discuss the data

with respect to potential evolutionary trajectories of A. gracilipes. As suggested by previous

studies, NE-Borneo is inhabited by a mosaic of variably related, ecologically dominant A.

gracilipes supercolonies (Drescher et al. 2007; Drescher et al. 2011) with unusually high

intracolonial relatedness estimates compared to all other supercolonial ant species studied so

far (Helanterä et al. 2009; Drescher et al. 2010). Thus, assuming that the scenario described

above applies to A. gracilipes, we expect that the varying genetic differentiation between

supercolonies corresponds to the degree of CHC-profile differentiation. As intranidal mating

may be the dominant, if not exclusive, reproductive strategy in A. gracilipes, we furthermore

expect that CHC-profiles differ both quantitatively as well as qualitatively between

supercolonies, and that qualitative differences between the CHC-profiles of supercolonies are


40

Figure 9. Location of six Anoplolepis

gracilipes supercolonies in Sabah, Malaysia. A. Location of the six Anoplolepis gracilipes colonies in the study area. B. Location of Poring Hot Springs in NE-Borneo. Colour codes correspond to colony affiliation and to the results of a Bayesian clustering algorithm under the assumption of K=4 genetic clusters (Fig. 12).

independent of substance classes as would be expected by an accumulation of random,

supercolony-specific mutations.

METHODS

Selection and maintenance of colonies

We localized six Anoplolepis gracilipes supercolonies within the study area around Poring

Hot Springs, Sabah, Malaysia (6°04’ N, 116°70’ E, colonies are from here on referred to as

supercolonies P1-P6, Figure 9). All supercolonies included in this study were at least 200m in

diameter, except for one slightly smaller supercolony (P1), which spanned only 100m (Fig.

9). The supercolonies were between 150 m (P1-P2) to ca. 15 km (P6-P5) apart. Workers,

brood and queens from a single nest per supercolony were collected and transferred into

plastic containers (45 × 25 × 25 cm) treated with Fluon™ to prevent escape (henceforth

termed subcolony). All subcolonies were offered newspaper as nesting material and were fed

water, honey and tuna every third day. Each of the subcolonies contained at least 1500

workers, 200 pupae/larvae and 3-12 queens (except for P1, which contained about 500

workers, 100 pupae and two queens).


41

Behavioural assays

Behavioural assays were performed within and between each of the six subcolonies (15

intercolonial combinations) using two different indices (Aggression Index AI and Mortality

Index MI). Both indices were measured by placing 5 individuals of each colony (10 ants per

trial per pairwise colony combination) inside a Fluon™ coated plastic cylinder (diameter=10

cm, height=5 cm) on a sheet of paper which was replaced after each trial. The average

maximum level of aggression (henceforth termed aggression index AI) was obtained as the

average of the most aggressive interaction within 5 minutes across ten replicates according to

the following categories: 1 – ants displayed no reaction towards each other/no physical

contact, 2 – reciprocal antennation, 3 – ants biting and spread-eagling one another, 4 – ants

protruding gaster/ spraying formic acid while biting opponent. Aggression levels 1 and 2 were

considered nonaggressive while aggression levels 3 and 4 were considered

antagonistic/aggressive. In order to measure the Mortality Index MI (as described in

(Drescher et al. 2007)), encounters between five workers of each subcolony were observed for

a period of 60 minutes. Every minute, the number of dead individuals was counted. For each

trial, the mortality index MI was obtained as MI = (y/2)/t50, with y being the total number of

individuals killed at the end of the experiment (60 min) and t50 the time (in steps of 1 min)

when half of this number (y/2) was already killed. Thus, this index allowed us to describe

aggression by combining both the number of dead individuals as well as the speed at which

ants kill each other.

Additionally, we measured aggression of workers towards allocolonial sexuals, i.e. virgin

alate queens and males. As not enough virgin queens and males could be collected from the

supercolonies in Poring Hot Springs, these tests were performed within and between three

supercolonies from the Sepilok Forest Reserve (supercolonies S6, S7 and S10, 5°52’ N,

117°57’ E) and one from the Lower Kinabatangan Nature Reserve (supercolony K1, 5°29’ N,

118°16’ E), Sabah, Malaysia. Aggression towards allocolonial sexuals in a combination of

supercolonies was measured by placing five workers from one supercolony and either one

alate queen or one male from another supercolony in an arena and vice versa. We then scored

presence/absence of aggression (biting, spread-eagling of the queen/male, protruding gaster

and spraying acid) in 10 trials of 60 min each. Overall aggression towards allocolonial virgin

queens/males was obtained by pooling the data from the 20 replicates per reproductive caste

per supercolony combination. Aggression between workers of the same supercolonies was


42

measured similarly, only that five individuals of each supercolony were placed in the arena

(N=10 replicates per supercolony combination).

Relatedness and population structure

Workers from each supercolony were sampled in 98.8% EtOH p.A. 20 workers from each of

the six subcolonies were genotyped using six polymorphic microsatellite loci (Ano1, Ano3,

Ano4, Ano6, Ano8, Ano10) according to the protocol in Feldhaar et al. (2006). Genetic

diversity within supercolonies was measured by genotyping five workers of two to four

additional nests per supercolony (two additional nests for P1 and P5, four nests for P2, P3, P4

and P6). This resulted in a total number of 220 genotyped individuals.

Relatedness within and between the six supercolonies was calculated using Relatedness 5.0.8

(Queller and Goodnight 1989), including all individuals as reference population. All R values

presented in this study arise from the half-matrix resulting from 220 × 220 pairwise

comparisons of individual relatedness excluding comparisons of each individual with itself.

Secondly, we conducted a three-level hierarchical analysis of molecular variance (AMOVA)

over the entire data set to determine how genetic variability was distributed across three levels

(between individuals within nests, among nests within a supercolony, among supercolonies)

in the population using ARLEQUIN 3.01 (Excoffier et al. 2005). Thirdly, we performed

assignment tests using STRUCTURE 2.2 (Pritchard et al. 2000; Falush et al. 2003, 2007) to

determine to what extent the patterns obtained with aggression tests correlated with genetic

differentiation between supercolonies. This software infers the number of clusters (K) that

best fits a data set by maximizing Hardy-Weinberg equilibrium and calculates the assignment

probability for each individual under any assumed K. We performed ten independent runs for

K from one to six. All genotyped individuals of a supercolony were included in this analysis.

Cuticular Hydrocarbon (CHC) Profiles

Cuticular hydrocarbons were extracted from 20 pooled workers from the same nests as those

sampled for bioassays (subcolonies) and genetic analysis plus two additional nests for

supercolony P2. Single workers did no yield sufficient amounts of substances to allow

analysis of the CHC profiles. The individuals were frozen at -20°C for 30 min prior to 8 min

extraction in hexane. Extracts were reduced under a gentle stream of nitrogen to 10 to 15 µl

and immediately used for analysis or stored at -20°C. 1 µl of the extract was analysed by gas

chromatography-mass spectrometry (GC-MS) using a Hewlett-Packard HP 6890 gas


43

chromatograph (GC, equipped with a J & W DB-5 fused silica capillary column: 30 m x 0.25

mm ID; film thickness: 0.25 µm) coupled with a Hewlett-Packard HP 5973 mass selective

detector (Hewlett-Packard, Waldbronn, Germany). We used a temperature program starting

from 60°C with an increase of 5°C/min until a final temperature of 300°C, which was kept for

10 min. A split/splitless injector was set to splitless mode for 60 sec at a temperature of

250°C. Helium was used as carrier gas with a constant flow of 1 ml/min. Electron ionization

mass spectra (EI-MS) were recorded at 70 eV with a source temperature of 230°C.

Only hydrocarbons, which were identified by their typical mass spectra, were included in our

analysis. Furthermore, all molecules smaller than C19-bodies were discarded, as they only

occurred in traces and were absent in most ant CHC profiles (Martin and Drijfhout 2009). We

compared differences in cuticular hydrocarbon profiles between supercolonies by performing

permutation tests (adonis, R-package vegan 1.15, 10000 runs) based on relative peak areas

(proportion of each peak’s integrated peak area to total integrated peak area). To visualize

differences in CHC profiles, we performed non-metric multidimensional scaling (NMDS)

based on Bray-Curtis dissimilarities (dij) of relative peak areas of CHC profiles.

Correlations between behaviour, genetic and chemical properties of workers from different

supercolonies

We constructed a matrix of the proportion of alleles that were not shared by supercolony pairs

and, likewise, a matrix of the proportion of qualitatively differing cuticular hydrocarbon

substances between supercolony pairs. We then tested for correlation between these two

matrices and the matrices of pairwise aggression (AI, MI), relatedness (R) and Bray-Curtis-

distances of CHC profiles (dij) by performing mantel tests (10000 permutations) between

matrix pairs. If not stated otherwise, all statistical analyses were performed with Statistica 7.1

(StatSoft, 2005) or R 2.9.2 (Ithaka and Gentleman 1996).

RESULTS

Behavioural assays

Both types of behavioural assays (AI, MI) yielded consistent results (R=0.65, p=0.002, Mantel

test). Aggression differed among supercolonies: While workers from P1-P2 and P3-P4

tolerated each other, workers from the remaining pairwise bioassays were aggressive towards


44

Figure 10. Aggression between six Anoplolepis gracilipes supercolonies from Sabah, Malaysia. Aggression was measured as (A) Aggression Index AI (mean ± SD, range: 1 - 4) and (B) Mortality Index MI (mean ± SD). The line widths are equivalent to the degree of aggression and dotted lines represent absent aggression. Colour codes correspond to colony affiliation and to the results of a Bayesian clustering algorithm under the assumption of K=4 genetic clusters (Fig. 12).

Table 2. Aggression of workers towards allocolonial workers, males and queens. The number of crosses represent the frequency of aggression of workers towards allocolonial queens (red crosses), males (blue crosses) and workers (black crosses) from four different Anoplolepis gracilipes supercolonies (supercolonies K1, S6, S7and S10; ++: aggression was observed in more than 50% of all replicates; +++: aggression observed in more than 75% of all replicates).

each other (Kruskal-Wallis ANOVA, AI: H = 177.5, p < 0.0001; MI: H = 144.4, p < 0.0001;

Figure 10). Assuming that P1-P2 and P3-P4 belonged to the same supercolony, despite being

separated up to ca. 1.5 km (P3-P4), we performed the statistical analysis again without colony

pairs P1-P2 and P3-P4. Still, aggression differed among the remaining pairs (AI: H = 36.1, p =

0.0003; MI: H = 60.4, p < 0.0001; Fig. 10). Furthermore, we observed high levels of

aggression of workers towards allocolonial sexuals (virgin queens/males) in at least 50% of

all replicates in each pairwise supercolony combination (Table 2).

K1 S6 S7 S10

K1 --- +++/+++ ++/++ ++/++

S6 +++ --- ++/++ +++/++

S7 ++ ++ --- ++/++

S10 ++ +++ +++ ---


45

Figure 11. Relatedness within and between six Anoplolepis gracilipes supercolonies from Sabah, Malaysia. Relatedness was calculated as R (mean ± SE, range: -1 – 1). The line widths are equivalent to the degree of relatedness. Colour codes correspond to colony affiliation and to the results of a Bayesian clustering algorithm under the assumption of K=4 genetic clusters (Fig. 12).

Relatedness and population structure

We found 29 alleles in the entire population (215 individuals from six spatially separated

colonies genotyped at six microsatellite loci). The number of alleles per locus ranged from

three to 11, while the number of alleles per locus per supercolony ranged between two and

four (Table 3). Laboratory subcolonies contained the same alleles as the other genotyped nests

from their respective supercolony. Mutually aggressive subcolony pairs shared less than half

of their alleles (33.1 ± 17.5%, mean ± SD), while mutually tolerant subcolonies P1-P2 and

P3-P4 shared either all alleles (P1-P2) or 85.7% (P3-P4, 16 of 18 alleles in common) of the

pairwise allele pool (Table 4). Workers within subcolonies were closely related (R = 0.85 ±

0.09, mean ± SE, range: 0.74 — 0.98, Fig. 11), while relatedness between reciprocally

aggressive subcolony pairs was low (R = -0.31 ± 0.47, mean ± SE, range: -0.69 — 0.51). The

mean relatedness between the two tolerant supercolony pairs P1-P3 and P3-P4 was as high as

intracolonial relatedness (Fig. 11). Accordingly, the genetic diversity in the population sample

originated from differences between supercolonies (AMOVA, FSCOLONY-TOTAL = 0.247, Table

5) rather than differences between

individuals in nests (FIND-NEST = -0.925) or

nests within supercolonies (FNEST-SCOLONY = -

0.002). Without a priori information on the

origin of the sampled individuals, the

Bayesian clustering algorithm implemented

in STRUCTURE 2.2 revealed the highest

likelihood of the data (Ln P(D)) between

K=4 and K=6, with K=4 showing the

smallest standard deviation (Fig. 12 A).

When using a priori information on assumed

numbers of K, the assignment of all

genotyped individuals to K=4 genetic

clusters was clearest, especially for

individuals from colonies P3 and P4 (Fig. 12

B).


46

Table 3 Alleles of six microsatellite loci in six Anoplolepis gracilipes supercolonies. The alleles of six microsatellite loci (Ano1, Ano3, Ano4, Ano6, Ano8, Ano10) are given in rows, the six Anoplolepis gracilipes supercolonies (P1-P6) and the entire study area (Total) are given in columns. Nn indicates the number of nests and Nw the number of workers genotyped per supercolony.

P1 P2 P3 P4 P5 P6 Total

Ano1

99

103

99 103

99 103

99 103

99 103

99 105

99 103 105

Ano3

164 166

164 166

144 156

144 156

144 156

164 166 168

144 156 164 166 168

Ano4

159 175

157 175

157 165

157 165

157 165

159 175

157 159 165 175

Ano6

116 130

116 130

116 130

116 130

116 118 130

116 130

116 118 130

Ano8

214 224 294

214 224 294

220 274 278

212 220 278

208 212 260 274

220 286 292

208 212 214 220 224 260 274 278 286 292 294

Ano10

234 242

234 242

234 238

234 238

234 238

234 242

234 238 242

Nn 3 5 5 5 3 5 26

Nw 30 40 40 40 30 40 220


47

Table 4. Percentage of alleles differing between Anoplolepis gracilipes supercolonies in relation to the pairwise allele pool.

Table 5. Multi-level genetic analysis of molecular variance (AMOVA) of six Anoplolepis gracilipes supercolonies. *: p < 0.05; **: p<0.01; ***: p < 0.001. Significance values were obtained by bootstrapping 10000 times over loci.

P1 P2 P3 P4 P5

P2 0

P3 76.2 76.2

P4 76.2 76.2 14.3

P5 78.3 78.3 35.3 35.3

P6 50 50 77.3 77.3 84

FIND-NEST FNEST-

SCOLONY

FSCOLONY-TOTAL

Ano1 -1 0 0.141*** Ano3 -0.923 -0.003 0.361*** Ano4 -1 0 0.376*** Ano6 -0.945 -0.001 0.009* Ano8 -0.712 -0.009 0.373*** Ano10 -1 0 0.231*** All -0.925 -0.002 0.247***


In total, we scored 154 peaks from GC-MS chromatograms of 26 A. gracilipes nest sites in

Poring Hot Springs. The chemical profiles of pooled workers of reciprocally tolerant

subcolonies were much more similar (Bray-Curtis distances dP1-P2 = 0.12 and dP3-P4 = 0.14,

respectively) than those of mutually aggressive subcolonies (dij = 0.32 ± 0.06, mean ± SD,

range: 0.22 – 0.42, Figure 13). Likewise, the cuticular hydrocarbon profiles from additional

nests of the six Anoplolepis gracilipes supercolonies in the field clustered according to

supercolony affiliation (ADONIS, F = 10.78, p< 0.0001, Fig. 14) and were highly similar

between the nests of supercolonies of which subcolonies were mutually tolerant (ADONIS,

P1-P2: F = 2.84, p = 0.18; P3-P4: F = 1.80, p = 0.23).


48

Figure 12. Estimated number of genetic clusters and assignment probability of six A. gracilipes supercolonies. (A) Average of the logarithm of the likelihood (Ln P(D)) of the data to be assigned to K genetic clusters as calculated with STRUCTURE 2.2. (B) Assignment probability of N=220 Anoplolepis

gracilipes workers from colonies P1 — P6 to K=4, K=5 and K=6 genetic clusters. Identical colours indicate that subgroups belong to the same genetic cluster. At K=5 and K=6, individuals from colonies P3 and P4 can not be assigned to a specific cluster. At K=4, individuals from colonies P1 and P2 are assigned to the red cluster, individuals from colonies P3 and P4 are assigned to the blue cluster and individuals from colonies P5 and P6 are each assigned an individual cluster.

Figure 13. Bray-Curtis dissimilarities of relative peak areas of cuticular hydrocarbon profiles between six Anoplolepis gracilipes supercolonies. Colour codes correspond to colony affiliation and to the results of a Bayesian clustering algorithm under the assumption of K=4 genetic clusters (Fig. 12).

Figure 14. NMDS plot of Bray-Curtis dissimilarities of CHC profiles between six A. gracilipes colonies. All symbols are coloured according to colony affiliation and the assignment to common genetic clusters (Fig. 12). Colony profiles differ between the four clusters (ADONIS, F=10.78, p=0.0001), but not between colonies P1-P2 (F = 2.84, p = 0.11) and P3-P4 (F = 1.80, p = 0.20).

We then limited the array of analyzed peaks

to only those, which constituted at least 1%

of relative peak area in at least one sample,


49

Figure 15. NMDS plots of Bray-Curtis dissimilarities of CHC profiles of six A. gracilipes supercolonies. We calculated Bray-Curtis dissimilarities of the CHC profiles using four different scoring thresholds (A-D). (A) Relative peak areas (RPA) were larger than 1‰ in at least one sample. (B) RPA’s larger than 5‰ in at least one sample. (C) RPA’s larger than 1% in at least one sample. (D) RPA’s larger than 2% in at least one sample. Colour codes correspond to colony affiliation and to the results of a Bayesian clustering algorithm under the assumption of K=4 genetic clusters (Fig. 12). While adonis results and stress values of the NMDS plots remained similar regardless of the predefined thresholds, the arrangements of NMDS plots were similar to the original arrangement at thresholds >1‰, >5‰ and >1% (A-C), but not at >2% (D).

thus reducing the array of cuticular compounds from 154 (all substances) to 30. Nevertheless,

ADONIS results and NMDS arrangements remained largely the same (Fig. 15). The

remaining 30 substances comprised a mix of unbranched and methyl-branched alkanes and

alkenes, 16 of which differed qualitatively between supercolonies (i.e. compounds that were

present in CHC profiles of one or several supercolonies while being absent in others, Fig. 16).


50

Figure 16. GC-MS chromatogram of a pooled sample of 20 Anoplolepis gracilipes workers from supercolony P3. 30 cuticular substances remained after applying a 1% scoring threshold. Green labels indicate cuticular compounds that are present in this sample, whereas red labels indicate substances that are absent from this sample, but present in cuticular extracts from other supercolonies. Asterisks in the compound list indicate the 16 substances that differ qualitatively between supercolony CHC-profiles.

Qualitative differences between CHC profiles involved all identified compound-classes, but

were overrepresented in dimethyl-branched alkanes (Fig. 16). Qualitatively differing

compounds constituted between 1.01% and 8.21% per compound of the respective CHC

profiles (average: 2.59 % ± 0.5 %, mean ± SE). Moreover mutually aggressive colonies

shared only 73.8% of the compounds in their CHC profiles on average, while mutually

tolerant colonies (P1-P2, P3-P4) shared 95.8% of the compounds (presence/absence of

compounds, Table 6).


supercolonies

The behavioural patterns among subcolonies were in perfect agreement with intercolonial

relatedness and Bray-Curtis similarities of CHC profiles. Intercolonial aggression (in both AI

and MI) was negatively correlated to relatedness (AI~R: r = -0.68, p = 0.023; MI~R: r = -0.84,

p = 0.002, Mantel test, Table 7) and positively correlated to Bray-Curtis distances of the CHC

profiles (AI~dij: r = 0.72, p = 0.015; MI~R: r = 0.62, p = 0.023). Thus, high aggression


51

Table 6. Percentage of Cuticular hydrocarbon compounds (CHC’s) differing between six Anoplolepis

gracilipes supercolonies in relation to the pairwise compound pool.

Table 7 Correlation coefficients between aggression, relatedness and chemical and genetic differentiation between six Anoplolepis gracilipes supercolonies. Abbreviations represent aggression (AI, MI), relatedness (R), Bray-Curtis distances of cuticular hydrocarbon profiles (dij CHC), percentage of distinct alleles (% Alleles) and percentage of distinct CHC compounds (% CHC) between six Anoplolepis gracilipes supercolonies. Asterisks indicate p-levels: +p<0.1, *p<0.05, **p<0.01.

occurred at low relatedness and high CHC profile dissimilarity, while low aggression matched

high relatedness and low dissimilarity of CHC profiles. Accordingly, relatedness and CHC

profile dissimilarity were negatively correlated (r = -0.77, p = 0.007). Moreover, the

proportion of alleles that differed between supercolonies pairs was positively correlated to the

proportion of CHC compounds that differed between supercolony profiles (r = 0.26, p =

0.044).

AI MI R dij CHC % Alleles

MI 0.65**

R -0.68* -0.84**

dij CHC 0.72* 0.62* -0.77**

% Alleles 0.85** 0.45+ -0.36+ 0.50+

% CHC’s 0.38* 0.05 0.09 0.25 0.26*

P1 P2 P3 P4 P5

P2 4.2

P3 12 8

P4 8.3 12 4.2

P5 37.9 34.5 34.5 37.9

P6 33.3 29.6 29.6 33.3 29.6


52

DISCUSSION

The present study revealed a significant positive correlation of aggression with differentiation

of the cuticular hydrocarbon profile as well as genetic differentiation among sympatric

supercolonies of the Yellow Crazy Ant Anoplolepis gracilipes in a study site in NE-Borneo.

Intercolonial aggression varied among supercolony pairs, ranging from tolerance to mortal

encounters in the bioassays. Mutually aggressive supercolonies were distantly related and

showed highly differentiated CHC profiles, while two non-aggressive though spatially

separated supercolonies were found to have highly similar CHC profiles and relatedness

values that were comparable to those found within supercolonies. Similar patterns have been

observed in the Argentine Ant Linepithema humile, where mutually tolerant supercolonies

from different locations (even from different continents) can be highly similar in genetic and

chemical terms, and are thus thought to have been introduced from the same source

population (Brandt et al. 2009b; Sunamura et al. 2009a; van Wilgenburg et al. 2010).

Likewise, supercolonies that are highly differentiated in genetic and chemical terms (even if

they are close to each other geographically) are thought to have been introduced from

different source populations (Sunamura et al. 2009b; Vogel et al. 2010). Hence, the varying

degrees of behaviour as well as genetic and chemical differentiation that we observed in A.

gracilipes in this study may potentially be explained by the same mechanisms, i.e.

supercolony pairs P1-P2 and P3-P4 may have been introduced from the same source

population whereas supercolonies P5 and P6 may have been introduced from yet another

population, respectively. Alternatively, two tolerant supercolonies may belong to the same

supercolony separated by uninhabitable terrain (very likely for the pair P1-P2 which was

separated by only 150m) or one supercolony in a tolerant supercolony pair may have

originated from the other through via human-mediated jump-dispersal within the study site

(possibly true for P3-P4).

Cuticular hydrocarbon (CHC) profiles have repeatedly been shown to enable ants to

discriminate between nestmates and non-nestmates (Vander Meer and Morel 1998; Lahav et

al. 1999; Akino et al. 2004). Typically, differing CHC profiles lead to aggression while

similar CHC profiles do not. Aggressive responses towards individuals bearing allocolonial

CHC profiles may be evoked according to a threshold rule (e.g. Cataglyphis iberica) or may

be gradual (e.g. Myrmica rubra) (reviewed in (Lenoir et al. 1999)). The positive correlation

between CHC profile dissimilarity and aggression that we observed between individuals from


53

different supercolonies in A. gracilipes may suggest the existence of a graded response

mechanism, which has also been found in other invasive ant species such as Lasius neglectus

(Ugelvig et al. 2008), Wasmannia auropunctata (Errard et al. 2005) and Linepithema humile

(Suarez et al. 2002; Torres et al. 2007; Blight et al. 2010).

Supercolonies of A. gracilipes showed both quantitative and qualitative differences in their

CHC profiles across various types of hydrocarbons (Fig. 4, Table 6). This is remarkable since

in most ant species, CHC profiles tend to differ quantitatively between conspecific colonies

and qualitatively between species (Vander Meer and Morel 1998; Elmes et al. 2002; Hefetz

2007). For instance, 14 out of 41 epicuticular compounds differed qualitatively between the

CHC profiles of six Formica species (Formica sensu stricto) in a study by Martin et al.

(2008b), if the same threshold was applied as in our study. In comparison, 16 out of 30

substances differed qualitatively between the profiles of six different supercolonies of A.

gracilipes in our study. On occasion, high degrees of qualitative differences between CHC

profiles of colonies from non-invasive ant species have been reported, but analyses either

include all substances (including trace substances, e.g. (Elmes et al. 2002)) or qualitative

differences are only detected in one substance class, (e.g. homologous series of (Z)-9-alkenes

or positional isomers of dimethyl-branched C25-bodies, (Martin et al. 2008a)). In invasive ant

species, qualitative differences between CHC profiles of different supercolonies have

repeatedly been found, e.g. between invasive and native populations of Linepithema humile

(Brandt et al. 2009b) and Wasmannia auropunctata (Errard et al. 2005). CHC profiles were

found to be less complex in introduced areas, with substances having been lost from the

profiles in comparison to colonies in the native range (Errard et al. 2005). However,

supercolonies of Linepithema humile also differ in their CHC profiles in the invasive range

(Blight et al. 2010).

Here, we found unusually high degrees of qualitative differences between CHC profiles of

different supercolonies of A. gracilipes. We identified 30 major CHC compounds that

accounted for at least 1% of the entire CHC profile and found qualitative differences in

compounds that accounted for up to 8.21% of the profiles of different A. gracilipes

supercolonies. The qualitatively differing substances comprised different compound classes

such as alkanes, alkenes, methyl-branched alkenes as well as mono-, di- and trimethyl-

branched alkanes of varying chain length (Fig. 4), as would be expected if supercolonies

diverge in a drift-like process of neutral evolution.


54

In our study, genetic distances corresponded to chemical distances between colonies, as has

been observed in several other studies on various social insect species (ants (Dronnet et al.

2006), termites (Dapporto et al. 2009) or paper wasps (Ugelvig et al. 2008; Vogel et al.

2009)). The correlation between differentiation in CHC profile and differentiation in

microsatellite loci in A. gracilipes suggests that cuticular hydrocarbons are to a large extent

genetically determined in this species (as is in many ant species (Adams 1991; Pennanec'h et

al. 1997; Takahashi et al. 2001; van Zweden et al. 2009; van Zweden et al. 2010)), albeit

environmental cues may also be involved (Liang and Silverman 2000; Florane et al. 2004).

Helanterä et al. suggested that unicoloniality (the ability of ant species to form populations

consisting of one or more supercolonies (Pedersen et al. 2006; Helanterä et al. 2009)), might

be an evolutionary dead-end (Helanterä et al. 2009), since selfish behaviour of distantly

related nestmates should lead to supercolony instability as predicted by kin-selection theory.

This view is supported by the scattered distribution of unicoloniality in the ant phylogeny as

well as the fact that no unicolonial ant species has a sister species that is also unicolonial

(Helanterä et al. 2009). In contrast to other supercolonial ant species such as L. humile and S.

invicta, however, workers in A. gracilipes supercolonies are closely related (Drescher et al.

2007; Helanterä et al. 2009), and thus selfish behaviour is less likely to evolve.

Despite this potential evolutionary dead-end, the limited or even absent gene flow between

supercolonies of unicolonial ant species (Jaquiery et al. 2005; Pedersen et al. 2006; Thomas et

al. 2006; Thomas et al. 2010b) may be conducive to speciation (Helanterä et al. 2009). As

speciation occurs on timescales beyond reach for current research, indirect approaches are the

only way to find clues regarding that question, e.g. by measuring how much supercolonies

have diverged in neutral or selected traits (Helanterä et al. 2009). As we have shown, A.

gracilipes supercolonies are indeed differentiated in neutral traits, i.e. alleles of microsatellite

loci and several cuticular compounds. Regardless of the origin of this initial differentiation

that we see today, one should expect both the genetic and chemical differentiation to increase

in the prolonged absence of gene flow between supercolonies. Although we cannot infer

absence of gene flow directly, the sociogenetic structure of the supercolonies in NE-Borneo

resembles that of the supercolonies found on Christmas Island, where gene flow among the

two supercolonies can be excluded based on mitochondrial and microsatellite markers

(Thomas et al. 2010b). In addition, behavioural characteristics of A. gracilipes, such as

intranidal mating soon after eclosion of female reproductives (pers. obs.) and aggression


55

towards allocolonial sexuals by workers (Table 2), suggest that mating is non-random,

resulting in strongly reduced or absent gene flow between supercolonies. Similar patterns

(intranidal mating, limited or absent intercolonial gene flow) have been reported for the

likewise invasive ant Linepithema humile both in its native (Pedersen et al. 2006; Vogel et al.

2009) and invasive range (Giraud et al. 2002; Jaquiery et al. 2005; Thomas et al. 2006).

The initial differences that would be necessary in order for subsequent differentiation between

neighbouring supercolonies to occur are likely to accumulate in allopatry, e.g. if

supercolonies stem from different source populations or if supercolonies are divided into two

or more fragments through human-mediated jump dispersal. In the latter scenario, lack of

contact zones (and thus mating) between the resulting allopatric fragments is likely to lead to

accumulation of colony (or fragment-) specific mutations and CHC compounds. If both

fragments come into secondary contact again, genetic and chemical differentiation may have

advanced enough to result in mutual aggression and lack of gene flow. This pattern was very

recently described for Linepithema humile supercolonies from Corsica and the European

mainland, where the introduction of one or several L. humile propagules from the mainland to

Corsica likely entailed an interruption of gene flow between Corsican and European

supercolonies, resulting in the pronounced chemical and behavioural differentiation

observable today (Blight et al. 2010). This scenario may also explain parts of our data, as

supercolonies P3 and P4 are likely to be established fragments of the same supercolony which

have already accumulated supercolony-specific alleles and cuticular compounds despite still

tolerating each other (Tables 4 and 6). Over time, the spatial separation (allopatry) of the two

supercolonies P3 and P4 should facilitate increasing genetic and chemical differentiation,

which in turn should lead to mutual aggression as soon as CHC profiles are sufficiently

different. Furthermore, genetic and chemical differentiation should also entail behavioural

suppression of intercolonial gene flow through worker aggression against allocolonial sexuals

even if they came into contact again (e.g. as a result of range expansion).

Thus, we suggest that the combination of exclusive intranidal mating and budding on the one

hand and positive feedback between genetic, chemical and behavioural traits on the other

hand may drive supercolonies towards ever increasing differentiation, possibly even involving

reproductive isolation and thus, speciation. In contrast to the majority of ant species, A.

gracilipes supercolonies are characterized by a combination of traits (polygyny, intranidal

mating, and lack of active dispersal other than budding) that ensure the continuous production


56

of generation upon generation of reproductives that all stay within the supercolony. Coupled

with an unusual reproductive system that may potentially avoid the negative effects of

inbreeding depression (Drescher et al. 2007), the extreme polygyny and polydomy of A.

gracilipes immensely reduces the risk of a breakdown of the entire supercolony, resulting in

virtual immortality of the supercolony superorganism. Furthermore, in eusocial species with

strict intranidal mating, a colony-specific signal should be sufficient for recognition between

the sexes and thus, selection towards a species-specific signal that allows recognition between

sexes from different supercolonies should be relaxed (as suggested in Martin et al. 2009). In

concert with the virtual immortality of A. gracilipes supercolonies, relaxed selection towards

a species-specific signal may even lead to a complete breakdown of intraspecific recognition

between sexuals from different colonies, potentially leading to prezygotic, reproductive

isolation. This, in turn, implies that different A. gracilipes supercolonies would no longer

belong to the same species according to the biological concept of species, which states that

species are groups of interbreeding natural populations that are reproductively isolated from

other such groups (Mayr 1942). Currently it is unclear whether different A. gracilipes

supercolonies should be perceived as belonging to the same species, as apparent lack of

random mating between sexuals from different supercolonies, intranidal mating and

aggression of workers towards allocolonial sexuals may already pose a sufficient barrier

preventing intercolonial gene flow. Thus, we argue that intercolonial differentiation may

potentially result in reproductive barriers between colonies, ultimately leading to independent

units on different evolutionary trajectories, and thus possibly speciation between

neighbouring A. gracilipes supercolonies.

ACKNOWLEDGEMENTS

We thank the Economic Planning Unit (EPU) for permission to conduct research in Malaysia,

Prof. Dr. Datin Maryati Mohamed for collaboration and Dr. Maklarin bin Lakim (Sabah

Parks) for permission to conduct our work at Poring Hot Springs, as well as H. Helanterä and

three anonymous reviewers for valuable comments on this manuscript. All experiments

conducted comply with the current laws of Malaysia.

VII. Worker aggression towards foreign sexuals

57

VII. WORKER AGGRESSION TOWARDS ALLOCOLONIAL MALES

AND QUEENS MAY FACILITATE SPECIATION BETWEEN

ANOPLOLEPIS GRACILIPES SUPERCOLONIES

ABSTRACT

Some ant species form large polygynous, polydomous supercolonies in which mating takes

place within the nest (intranidal mating) and which grow via the occupation of suitable

nesting sites by fertilized queens along the supercolony boundary (colony budding).

Supercolonial ant species rarely, if ever, perform nuptial mating flights and as a consequence,

gene flow may be largely confined within supercolonies whereas gene flow between

supercolonies may be limited or even absent. In the absence of mating flights, the migration

of reproductive individuals from one supercolony to another may represent the only way by

which gene flow between different supercolonies may be maintained. This, however, would

require workers to tolerate males or queens from different supercolonies. We tested the

potential of reproductive individuals to migrate between supercolonies using 14 supercolonies

of the Yellow Crazy Ant Anoplolepis gracilipes from North-East Borneo. A previous study

suggested that mutually aggressive supercolonies of this ant species may be highly

differentiated both genetically (in nuclear microsatellite loci) and chemically (in cuticular

hydrocarbons, CHC’s) suggesting limited or absent gene flow between them (Chapter VI,

Drescher et al. 2010). The present study revealed highly congruent patterns of genetic,

chemical and behavioural differentiation between A. gracilipes supercolonies, thus supporting

the findings of the previous study. Intercolonial differentiation in mitochondrial DNA also

corresponded to nuclear genetic, chemical and behavioural differentiation, implying that

maternal ancestry may be an important factor in shaping intercolonial differentiation. Arena

experiments showed that workers were highly aggressive towards males and queens from

different supercolonies, suggesting that they may act as behavioural boundary to a potential

migration of sexuals between supercolonies, thus possibly impairing gene flow between

supercolonies. A series of preliminary breeding experiments revealed that despite males and

queens from different supercolonies may be able to successfully mate in the absence of

workers, queens fertilized by allocolonial males may potentially not be able to produce viable

worker offspring. Thus, gene flow between different A. gracilipes supercolonies may not only


58

be impaired by worker aggression towards allocolonial reproductives, but may further be the

result of reproductive isolation between sexuals from different supercolonies. Both potential

mechanisms may enhance genetic and chemical differentiation between different A. gracilipes

supercolonies by impeding intercolonial gene flow, and might ultimately contribute to the

diversification of different A. gracilipes supercolonies into different species.

INTRODUCTION

Biological invasions are considered as one of the main threats to global biodiversity, as

invaders often negatively influence native taxa (e.g. Sanders et al. 2003; Kenis et al. 2009),

potentially leading to severe economic and environmental damage (Mack et al. 2000b;

Mooney and Cleland 2001; Perrings et al. 2002). Social insects, especially ants, belong to the

most devastating groups of invaders (Moller 1996) and five ant species are found among the

100 invasive species declared most destructive worldwide (Lowe et al. 2000): the Yellow

Crazy Ant (Anoplolepis gracilipes), the Argentine Ant (Linepithema humile), the Big-headed

Ant (Pheidole megacephala), the Little Fire Ant (Wasmannia auropunctata) and the Red

Imported Fire Ant (Solenopsis invicta). The ecological success of introduced ant species in

new habitats seems to be tightly linked to their unusual ability to form high-density

supercolonies, which are often polygynous (contain more than one queen) and in which

workers roam freely between neighboring nests (Holway et al. 2002; Pedersen et al. 2006). As

opposed to most ants, supercolonial ant species show a tendency to mate within the nest

(intranidal mating) instead of mating in periodic nuptial flights. Queens often stay within the

maternal nest after fertilization, enter neighboring nests within the supercolony or establish

nests at the supercolony boundary (known as supercolony budding). While the main dispersal

mode over short distances seems to be colony budding, the predominant long-range dispersal

mode of supercolonial ant species is human-mediated jump-dispersal of propagules (Holway

et al. 2002). Recent studies on L. humile (Jaquiery et al. 2005; Pedersen et al. 2006; Thomas

et al. 2006) and A. gracilipes (Drescher et al. 2010; Thomas et al. 2010a) suggest that as a

consequence of this combination of mating and dispersal strategies, gene flow may be largely

confined within supercolonies and thus extremely limited or even absent between them, thus

possibly resulting in ant supercolonies following independent evolutionary trajectories

(Drescher et al. 2010).


59

A vital prerequisite for supercolonies to diverge through absence of intercolonial gene flow is

that males and queens from different supercolonies do not mate. In most ant species, males

and queens from different colonies mate in synchronized nuptial mating flights. The absence

of nuptial mating flights, however, is a characteristic that is shared by several invasive ant

species such as Linepithema humile (Passera and Aron 1993; Aron 2001) or Lasius neglectus

(Espadaler and Rey 2001; Espadaler et al. 2004). Even though mating flights were suggested

for Anoplolepis gracilipes in a survey on Christmas Island (Abbott 2006), a later study

demonstrated that should these mating flights occur, they do not result in random mating

among sexuals from the different supercolonies on Christmas Island (Thomas et al. 2010a).

An alternative way by which males and queens could find allocolonial mates would be to

immigrate into foreign supercolonies. However, in order to facilitate such allocolonial mating

by migration of sexuals and hence gene flow between supercolonies, resident workers would

have to tolerate foreign males or queens entering their supercolony. Otherwise, mating

between males and queens from different colonies may be unlikely. Consequently, rejection

of allocolonial sexuals might represent an effective behavioural barrier to gene flow between

supercolonies, facilitating an independent co-existence and evolution of supercolonies.

In order to study whether the potential rejection of foreign sexuals by workers may act as

mechanism facilitating differentiation between supercolonies of Anoplolepis gracilipes, we

studied 14 A. gracilipes supercolonies in NE-Borneo using four different approaches

(behavioural assays, genetic analyses of mtDNA and microsatellites as well as chemical

analyses of cuticular hydrocarbon profiles). An earlier study suggested that aggression

between A. gracilipes supercolonies is positively correlated to their genetic distance, which is

reflected by a higher dissimilarity in their composition of cuticular hydrocarbons (Drescher et

al. 2010). This study further revealed that mutually aggressive A. gracilipes supercolonies

shared only a third of their alleles across six microsatellite loci and only three quarters of the

compounds in their cuticular hydrocarbon profiles (Drescher et al. 2010). Thus, we expect

aggression between workers from the 14 different supercolonies to be reflected by both the

genetic and chemical differentiation between them, and that mutually aggressive

supercolonies share only a fraction of their alleles and CHC compounds. Furthermore,

supercolonies of the same matriline (inferred by an identical mtDNA haplotype) might be

more closely related than supercolonies from different matrilines, since they may have been

derived from the same founding queen. Similar cuticular hydrocarbon profiles between such


60

Figure 17. Location of the 14 Anoplolepis gracilipes supercolonies used in this study. Ants were sampled from supercolonies in three regions in NE-Borneo: Poring Hot Springs (POR, three supercolonies), Sepilok Nature Reserve (SEP, ten Supercolonies) and Kinabatangan Nature Reserve (KIN, one colony).

closely related supercolonies may lead to lower levels of aggression, as opposed to encounters

between different haplotypes. Finally, high degrees of genetic and chemical differentiation

between mutually aggressive invasive ant supercolonies may suggest a lack of gene flow

between them (Drescher et al. 2010; Thomas et al. 2010a). We thus investigate whether the

immigration of allocolonial sexuals may be hampered by aggression of resident workers

towards them. We expect workers from mutually aggressive supercolonies to show a

corresponding aggression towards males and queens as well.

METHODS


Anoplolepis gracilipes subcolonies were collected and maintained from 14 supercolonies in

three focal regions in NE-Borneo (Fig. 17): Three from Poring Hot Springs (PHS,

supercolonies P3, P5, and P6, Drescher et al. 2010; Drescher et al. 2011), ten from the Sepilok

Forest Reserve (SEP, supercolonies S1-S10)

and one from the Lower Kinabatangan

Nature Reserve (KIN, supercolony K1).

Queens, workers and brood were collected

from each supercolony and transferred into

plastic buckets (Vol. = 25 l) treated with

Fluon™ to prevent escape. All colonies were

offered newspaper and cardboard as nesting

material and were fed water, honey and

crayfish scraps twice a week. Initially, each

of the subcolonies contained at least 10.000

workers, 500 pupae, 50 males and 20

queens.

Behavioural assays

Aggression between workers of the 14 subcolony pairs was obtained using two indices, the

Mortality Index MI (Drescher et al. 2007; Drescher et al. 2010) and the Aggression Latency

Tbite. To measure the Mortality Index MI, five workers of each subcolony (10 ants per trial per


61

pairwise colony combination) were placed within a Fluon coated plastic cylinder (diameter =

10 cm, height = 5 cm) on a sheet of paper which was replaced after each trial (n=10 trials per

colony combination). Every minute, the number of dead individuals was counted over a

period of 60 min. For each trial, the mortality index MI was obtained as MI = (y/2)/t50, with y

being the total number of individuals killed at the end of the experiment (60 min) and t50 the

time (in steps of 1 min) when half of this number (y/2) was already killed.

To measure the Aggression Latency Tbite, five ‘resident’ workers were confronted with one

allocolonial worker in an arena as described above. Within ten minutes, the Aggression

Latency Tbite was recorded as the time lapse from first antennal contact between ‘resident’

workers and the single allocolonial worker until the singe allocolonial worker was bitten for

the first time. We performed 20 trials for each subcolony combination, ten of which were

performed using ‘resident’ workers from one colony and the other ten using ‘resident’

workers from the other subcolony. Overall aggression towards allocolonial workers in a

subcolony combination was obtained by pooling the Aggression Latency Tbite across the two

sets of reciprocal bioassays. Likewise, aggression of workers towards allocolonial

reproductives was measured as the Aggression Latency Tbite among workers, males and

dealate queens from four different subcolonies (K1, S6, S7, S10). The Mortality Index MI was

not obtained for bioassays involving allocolonial workers and reproductives, as the lowered

aggression of workers towards allocolonial sexuals did not meet the requirements of the

Mortality Index MI.

Genetic analyses – mitochondrial DNA

As preliminary results yielded no more than one haplotypes per supercolonies (unpub.), one

individual each of the 14 supercolonies was used for DNA extraction using the Puregene®

DNA Purification Kit (Gentra Systems) according to the manufacturer’s recommendations.

Mitochondrial DNA was amplified using the primer pairs Horst (5’-

AC(TC)ATACTTTTAACTGATCG-3’) designed by D. Kronauer (unpubl.) / Ben (5’-

GC(AT)AC(AT)AC(AG)TAATA(GT)GTATCATG-3’) (Moreau et al. 2006) for partial

Cytochromeoxidase I (COI) (Genebank accession numbers: DQ888821-DQ888825)

corresponding to positions 2407-2427 (Horst) and 2891-2914 (BEN) relative to the

mitochondrial genome of Apis mellifera (Crozier and Crozier 1993) and CBI/CBII for partial

Cytochrome B (Cytb) (Genebank accession numbers: DQ888817-DQ888820) (Crozier et al.

1991). PCR reactions were performed as described in Drescher et al. (2007). The purified


62

mtDNA fragments were sequenced by SEQLAB, analyzed using Chromas Lite 2.01

(Technelysium Pty Ltd) and aligned using ClustalW (Thompson et al. 1994) implemented in

BioEdit 7.0.5.3 (Hall 1999). We then constructed a haplotype network based on the

concatenated sequence of 904bp (consisting of a 460bp fragment of the COI gene and a

444bp fragment of the Cytb) using the 95% parsimony algorithm implemented in TCS 1.21

(Clement et al. 2000a).

Genetic analyses – nuclear DNA / microsatellites

DNA was extracted from 8 individuals per subcolony as described above and genotyped using

six polymorphic microsatellite loci (Ano1, Ano3, Ano4, Ano6, Ano8, Ano10) according to

the protocol in Feldhaar et al. (2006). Relatedness within and between the 14 supercolonies

was calculated using Relatedness 5.0.8 (Queller and Goodnight 1989), including all

individuals as reference population. All R values in this study result from the half-matrix of

112 × 112 pairwise comparisons of individual relatedness, excluding comparisons of each

individual with itself.

Cuticular hydrocarbon profiles

Cuticular hydrocarbons (CHCs) were extracted from 20 pooled workers from each of the 14

supercolonies and analyzed according to the protocol in Drescher et al. (2010). Only

hydrocarbons, which were identified by their typical mass spectra, were included in our

analysis. Furthermore, all molecules smaller that C19-bodies were discarded since they only

occurred in traces and were absent in most of the investigated cuticular hydrocarbons (Martin

and Drijfhout 2009). To visualize differences in CHC profiles, we performed non-metric

multidimensional scaling (NMDS) based on Bray-Curtis dissimilarities (dij) of relative peak

areas of CHC profiles.


supercolonies

We constructed a matrix of the percentage of alleles that were not shared by supercolony pairs

as well as a matrix of the percentage of qualitatively differing cuticular hydrocarbon

compounds between supercolony pairs. We then tested for correlation between these two

matrices, the matrices of both aggression indices (Mortality Index MI and Aggression Latency

Tbite), relatedness R, Bray-Curtis-distances of CHC profiles and the genetic distance between


63

mtDNA haplotypes (i.e. the number of mutational steps needed to transform one haplotype

into another by substitution of single nucleotides) by performing mantel tests (10000

permutations) between matrix pairs.

Mating experiments between A. gracilipes supercolonies (preliminary)

Based on observations of worker aggression towards foreign sexuals, we examined whether

males and queens from different supercolonies were generally able to mate and start new

colonies, both in the presence and absence of workers from the queen supercolony. Virgin

queens were obtained by removing queen pupae from lab colonies and letting them hatch and

be cared for by their fellow workers in small subcolonies. In each of eight replicates, one

virgin alate queen was confronted with five males from a different lab colony in a plastic

container (10cm x 10cm, height = 5cm) which contained a small burrow made of plaster as

well as water, honey and bits of cockroaches. In four of the eight replicates, virgin queens

were accompanied by 150 workers from their own colony before five foreign males were

added. In the other four replicates, virgin queens and males were left alone for 72h after

which the five males were removed and 150 workers from the queen’s colony were added.

Every month, the presence/absence of all stages of brood was noted. After six months,

experiments were aborted and the queen’s spermathecae were dissected to verify whether the

queens had been inseminated. To compare colony development after intercolonial and

intracolonial mating, we removed 35 dealate queens from various supercolonies (thus already

mated queens), placed them in plastic containers as described above and added 150 workers.

The presence/absence of brood was noted as above. After three to six months, queens were

removed and their spermathecae dissected. As A. gracilipes supercolonies are highly

genetically differentiated (Drescher et al. 2010), the thorax and spermatheca content of each

queen was genotyped as described above to confirm that queens had been fertilized by males

from their own supercolony.

RESULTS

Behavioural, genetic and chemical differentiation between supercolonies

The intensity of aggression between allocolonial workers varied between supercolony pairs,

confirmed by both aggression indices (Kruskal-Wallis-ANOVA, MI: H = 559.8, p < 0.0001;


64

Table 8. Correlation coefficients between aggression and chemical and genetic differentiation among 14 Anoplolepis gracilipes supercolonies. Abbreviations stand for pairwise Mortality Index (‘MI’), Aggression Latency (‘Tbite’), Relatedness (‘R’), percentage of differing alleles (‘% Alleles’), Bray-Curtis distance of cuticular hydrocarbon profiles (‘dij CHC’), percentage of differing CHC compounds (‘% CHC’s’) and number of mtDNA mutational steps (‘mtDNA mutations’) between workers of 14 A. gracilipes supercolony pairs in Sabah, Malaysia. P-levels are: +p<0.1, *p<0.05, **p<0.01, ***<0.001, mantel tests.

Tbite: H = 589.3, p < 0.0001; Mantel test between MI and Tbite: R = -0.26, p = 0.0009, Table 8),

ranging from trials with no mortality to encounters in which 9 out of 10 ants died. Overall, 2.3

± 1.5 (mean ± SD) individuals died in bioassays between supercolonies, while not a single

individual was killed in intracolonial control tests. Workers of all supercolonies were

aggressive towards each other, and in 5:1 live assays, single workers were bitten after Tbite =

7.3 ± 5.5 sec (mean ± SD) after first contact with the ‘resident’ workers.

We found 32 alleles in the entire population. Relatedness between supercolonies was lower (R

= -0.04 ± 0.37, mean ± SD, range: -0.6 – 0.65) than within supercolonies (R = 0.81 ± 0.12,

range: 0.58 – 1; U = 62.0, p < 0.0001, Mann-Whitney U-test), and on average, supercolonies

shared 46.4% ± 16.4% of their alleles (mean ± SD). The GC-MS analyses of cuticular

hydrocarbons revealed a total of 129 cuticular compounds. Supercolonies shared 86.7% ±

5.3% of the compounds in their CHC profiles. Furthermore, intercolonial aggression, genetic

distance and qualitative chemical differentiation were significantly correlated between the

colony pairs, but quantitative distances of CHC profiles were only significantly correlated

with relatedness R (Table 8).

MI Tbite R

%

Alleles dij CHC

%

CHC’s

Tbite -0.26**

R -0.54*** 0.37***

% Alleles 0.55*** -0.30** -0.80***

dij CHC 0.28+ -0.08 -0.22* 0.10

% CHC’s 0.44*** -0.25* -0.61*** 0.51*** 0.15

mtDNA

mutations 0.19 -0.06 -0.15+ 0.23+ -0.18 0.29*


65

Figure 18. 95%-parsimony haplotype network based on a 904bp fragment of COI (460bp) and Cytb (444bp). Abbreviations denominate colony identity and sampling region (supercolonies P3, P5 and P6 from Poring/POR, supercolonies S1-S10 from Sepilok/SEP and supercolony K1 from Kinabatangan/KIN) that are distributed among 7 haplotypes (A-G). Nodes between haplotypes indicate the number of parsimonious mutational differences.

Figure 19. Aggression between Anoplolepis

gracilipes supercolonies of the same (‘within’) or differing (‘between’) mt-DNA haplotypes. Aggressoin was measured as (A) Mortality Index MI and (B) Aggression Latency Tbite (mean ± SD, including Mann-Whitney U-test statistics).

Behavioural, genetic and chemical differentiation between mtDNA haplotypes

We identified seven different mitochondrial haplotypes across the 14 A. gracilipes

supercolonies which were between 2 and 9 mutational steps apart (Fig. 18). The most

common haplotype A was found in five supercolonies from Sepilok whereas the other six

haplotypes were found in either one or two colonies. While six of the seven haplotypes were

exclusive to a single sampling region, haplotype C was found in Poring as well as in Sepilok.

The number of mutational steps that separated supercolony haplotypes was not correlated to

their aggression (MI, Tbite), genetic differentiation (R, percentage of differing alleles) or

quantitative dissimilarities between CHC profiles, but were positively correlated to the

percentage of distinct CHC compounds (R = 0.29, p = 0.017, Table 8, Mantel test). Workers

of all supercolonies displayed aggression towards each other. However, workers from the

same mtDNA haplotype were slightly less aggressive towards each other than workers from

supercolonies carrying a different haplotype (MIwithin = 3.8 ± 7.0, MIbetween = 9.2 ± 7.9, Fig. 19

A; Tbite-within = 8.6 ± 6.2 sec, Tbite-between = 10.1 ± 28.3, Fig. 19 B, mean ± SD). Likewise,

relatedness R was higher within a haplotype

than between haplotypes (Rwithin = 0.28 ±

0.35; Rbetween = -0.05 ± 0.38, Fig. 20 A) as was

allelic differentiation (percentage of


66

Figure 20. Genetic differentiation between Anoplolepis gracilipes supercolonies of the same (‘within’) or differing (‘between’) mt-DNA haplotypes. Genetic differentiation was measures as (A) Relatedness R and (B) percentage of differing alleles between supercolony pairs (mean ± SD, including Mann-Whitney U-test statistics).

Figure 21. Chemical differentiation between Anoplolepis gracilipes supercolonies of the same (‘within’) or differing (‘between’) mt-DNA haplotypes. Chemical differentiation was measured as (A) Bray-Curtis dissimilarities of CHC profiles and (B) percentage of compounds differing between CHC-profiles (mean ± SD, including Mann-Whitney U-test statistics).

different alleles within haplotypes: 44.1% ± 17.3%, between haplotypes: 55.1% ± 15.8%, Fig.

20 B). Whereas the quantitative chemical differentiation within and between haplotypes did

not differ significantly (Fig. 21 A), the qualitative chemical composition did (percentage of

different CHC compounds within haplotypes: 7.9% ± 4.3%; between haplotypes: 14.2% ±

4.9%, Fig. 21 B). The genetic distance between mtDNA haplotypes was positively correlated

to the quantitative chemical differentiation and marginally significantly correlated to nuclear

genetic differentiation (both R and proportion of differing alleles), but not to aggression or

quantitative distances between CHC-profiles (Table 8).

Worker aggression towards allocolonial males and queens

Workers were aggressive towards allocolonial workers as well as towards allocolonial males

and dealate queens (Fig. 22). However, workers attacked allocolonial workers much more

rapidly (Tbite = 9.2 ± 5.8 sec, mean ± 95% CI, Fig. 22 A) than allocolonial males (Tbite = 150.5

± 62.5 sec) or allocolonial queens (Tbite = 234.3 ± 50.0 sec). Furthermore, workers killed a

higher number of allocolonial workers and males than allocolonial queens (Fig. 22 B).

Worker aggression against allocolonial workers and males differed significantly among

supercolonies (aggression towards allocolonial workers: Tbite , H = 29.6, p < 0.0001;


67

aggression towards allocolonial males: Tbite, H = 15.8, p = 0.007, Kruskal-Wallis ANOVA,

Fig. 23 A-B), unlike aggression against allocolonial queens (Fig. 23 C). Nonetheless, worker

aggression towards allocolonial males and queens was positively correlated (Tbite, R = 0.75, p

= 0.042), but not between workers and males or workers and queens, respectively.

Mating experiments between A. gracilipes supercolonies (preliminary results)

In intercolonial breeding experiments in which males were added to a subcolony containing

150 workers and one virgin queen from a different colony, all males were instantly attacked

and killed within 30 minutes. Consequently, queens and males could not mate and no brood

was produced (see Table 9). In breeding experiments in which workers from the queen’s

colony were added 72 hours later than allocolonial males, all queens mated with one or

Figure 22. Aggression of Anoplolepis gracilipes workers towards allocolonial workers, males and queens. Aggression was measured as (A) Aggression Latency Tbite and (B) number of mortal encounters in 5:1 live bioassays (mean ± 95% CI, including Kruskal-Wallis ANOVA test statistics).

Figure 23. Aggression of workers from four Anoplolepis gracilipes supercolonies (K1, S6, S7, S10) towards (A) allocolonial workers, (B) allocolonial males and (C) allocolonial queens. Two aggression measures are displayed: Aggression Latency Tbite (black bars, white letters) and the number of mortal encounters in N=20 trials (grey bars, black letters; each trial consisted of 5 workers vs. 1 allocolonial worker/male/queen, mean ± 95% CI).


68

several of the allocolonial males, which was both observed as well as verified by the filled

spermathecae of all four queens. In two of the four replicates, no brood was produced despite

the fact that the queens were fertilized, and queens died after four months. In the remaining

two replicates, queens produced eggs and larvae, but no pupae were produced within six

months (Table 9). In contrast, all of the 35 dealate queens taken directly from within

supercolonies were inseminated (by males from their own supercolony) and started producing

pupae after a maximum of three months.

DISCUSSION

In an earlier study, we showed that supercolonies of the Yellow Crazy Ant Anoplolepis

gracilipes can be substantially differentiated both genetically (using nuclear DNA

microsatellites) and chemically to an extent that suggests lack of gene flow between them

(Drescher et al. 2010). In the present study, we (A) confirm the previous data with an

increased sample size of 14 supercolonies, (B) show that differentiation in mitochondrial

DNA corresponds to an increased aggression, nuclear genetic and chemical differentiation

between supercolonies and (C) suggest a mechanism by which gene flow may be suppressed

between supercolonies, i.e. the aggression of workers towards foreign sexuals. Moreover, we

queens/ worker colony

Males colony

Workers present?

Queens fertilized?

Colony development

S9 P5 YES NO all males died within 10min




S9 S1 NO YES 4th month: queen died, no brood during that time

S9 S1 NO YES 4th month: queen died, no brood during that time

S9 P5 NO YES 1st month: eggs, 3rd month: larvae, no pupae/callows until

abortion of the experiment

S9 P5 NO YES 1st month: eggs, 3rd month: larvae, no pupae/callows until

abortion of the experiment

Table S1 Development of Anoplolepis gracilipes subcolonies in allocolonial mating experiments under presence/absence of 150 workers from the queens’ colony within the first 72 hours of the experiment.


69

present preliminary data indicating that – in contrast to males and virgin queens from the

same supercolony – allocolonial reproductives may be able to mate but not necessarily

capable of producing viable worker offspring.

Behavioural, genetic and chemical differentiation between supercolonies

Recently, several studies have suggested limited or even absent gene flow between

supercolonies of supercolonial ant species such as Linepithema humile (Jaquiery et al. 2005;

Pedersen et al. 2006; Thomas et al. 2006) or Anoplolepis gracilipes (Drescher et al. 2010;

Thomas et al. 2010a), using behavioural and genetic or chemical properties (or both) as

indicators of differentiation between supercolonies. In this paper, we used two aggression

indices (Mortality Index MI and Aggression Latency Tbite), two measures of genetic

(Relatedness R and proportion of differing alleles) and two measures of chemical properties

(quantitative and qualitative distances of CHC profiles), all of which were correlated among

each other with the exception of quantitative CHC distance. By confirming previous data

(Drescher et al. 2010) in a second region containing a larger set of supercolonies, this study is

consistent with the hypothesis that the majority of A. gracilipes supercolonies existing today

may not exchange genetic material. Instead, they seem to represent different independent

units, i.e. potentially separated evolutionary trajectories.

Behavioural, genetic and chemical differentiation between mtDNA haplotypes

The 14 supercolonies in this study contained seven different mtDNA haplotypes which were

between two and nine mutational steps apart. Workers from all 14 supercolonies were

aggressive towards each other, which suggests that different haplotypes may virtually always

represent different supercolonies, while the same haplotype may define groups either from the

same or from different supercolonies (Drescher et al. 2010). Moreover, supercolonies of the

same haplotype may have been either introduced from the same source population or derived

from each other by dispersing queens (most likely as part of an anthropogenically dispersed

propagule). In contrast to data on the Argentine Ant Linepithema humile (Vogel et al. 2009),

A. gracilipes supercolonies with identical haplotype were genetically more similar that those

with different haplotypes, both in terms of pairwise relatedness as well as the percentage of

different alleles. Additionally, we observed less mortal aggression and a lower qualitative

chemical differentiation between supercolonies with identical haplotype. Thus, our data

suggest that ancestral maternal lineages may represent the first stage of differentiation


70

between A. gracilipes supercolonies followed by a second stage of differentiation within

maternal lineages, e.g. through accumulation of supercolony-specific nuclear genetic

mutations in allopatry (Drescher et al. 2010).

Aggression towards foreign sexuals and breeding experiments

As opposed to most social insects, supercolonial ant species such as Anoplolepis gracilipes or

Linepithema humile are characterized by a strong tendency to mate within the nest rather than

in synchronized nuptial mating flights. As a result, the immigration of reproductives into

foreign supercolonies may represent the only alternative possibility by which gene flow

between supercolonies may be achieved. A vital prerequisite for immigration of foreign

queens and males would thus be the tolerance by resident workers towards them. Our results,

however, indicate that a potential immigration of reproductives into foreign supercolonies

may consistently fail due to worker aggression towards allocolonial males and queens. Both

measures, the aggression latency Tbite and the number of lethal bioassays, yielded conclusive

evidence suggesting that whenever workers of different supercolonies are aggressive towards

each other, they are also likely to be aggressive towards each others reproductive castes.

Aggression differed to some extent between castes. While the workers’ aggression towards

foreign workers and males was both more rapid and more intense, it was significantly reduced

towards foreign queens. A higher tolerance towards dealate, fertilized queens may be due to a

fertility signal which has been shown to reduce worker aggression towards egg-layers in

several ponerine ant species such as Pachycondyla sublaevis (Ito and Higashi 1991),

Dinoponera quadriceps (Monnin and Peeters 1999), Diacamma ceylonense (Cuvillier-Hot et

al. 2002), and Streblognathus peetersi (Cuvillier-Hot et al. 2004).

Nevertheless, workers were aggressive towards both males and queens that did not belong to

the same supercolony, suggesting that a potential immigration of foreign reproductives may

likely be prevented by resident workers. This would effectively prohibit gene flow between

supercolonies. This view is further supported by preliminary data on mating experiments

between queens and males from different supercolonies (Table 9), which revealed that

workers fight off and kill foreign males even if they were in the presence of an unmated alate

queen. Moreover, our preliminary results indicate that the reproductive potential of A.

gracilipes queens fertilized by allocolonial males may be severely lowered, even though egg-

policing by the workers accompanying the queen can not be entirely excluded.


71

Overall, the present study is in concert with previous findings by Drescher et al. (2010),

demonstrating that neighboring Anoplolepis gracilipes supercolonies are strongly

differentiated both genetically and chemically, which translates into increased aggression

between supercolonies. In addition, the present study revealed that the affiliation to mtDNA

haplotypes corresponds to intercolonial aggression as well as nuclear genetic and chemical

differentiation, i.e. supercolonies that may stem from the same ancestral queen are slightly

less aggressive and genetically and chemically more similar than supercolonies from different

maternal lineages. Finally, the present study revealed a potential mechanism by which gene

flow between A. gracilipes supercolonies may be restricted or even entirely prohibited, i.e. the

aggression of workers towards allocolonial sexuals. As A. gracilipes supercolonies are

virtually immortal due to their capability to continuously produce generation after generation

of reproductives, a prolonged absence of gene flow may further enhance genetic and chemical

differentiation between supercolonies through neutral drift over long time spans. Positive

feedback between behavioural, genetic and chemical differentiation should further enhance

intercolonial segregation, possibly even until reproductive isolation, and thus speciation,

between different A. gracilipes supercolonies. In fact, the preliminary allocolonial mating

experiments may already have detected an intermediate form of reproductive isolation

between A. gracilipes supercolonies, as colony development was impeded in all four

replicates despite the fact that the queens were fertilized.

ACKNOWLEDGEMENTS


and Prof. Dr. Datin Maryati Mohamed for collaboration. Dr. Maklarin bin Lakim (Sabah

Parks) and Dr. Arthur Chung (Sabah Forestry Department) are thanked for permission to

conduct our study in Poring Hot Springs and Sepilok Forest Reserve, respectively. All

experiments conducted comply with the current laws of Malaysia.


72

VIII. Ecological dominance of Anoplolepis gracilipes

73

VIII. ECOLOGICAL DOMINANCE OF ANOPLOLEPIS GRACILIPES IN

AN ANT COMMUNITY IN NORTH-EAST BORNEO

This Chapter has been published as:

Drescher J, Feldhaar H, Blüthgen N (2011) Interspecific aggression and resource

monopolization of the invasive ant Anoplolepis gracilipes in Malaysian Borneo. Biotropica

43: 93-99

ABSTRACT

Invasions by introduced ant species can be ecologically destructive and affect a wide range of

taxa, particularly native ants. Invasive ant species often numerically dominate ant

communities and outperform native ant species in effective resource acquisition. Here we

describe interactions between the invasive ant Anoplolepis gracilipes (Smith) and resident ant

species in disturbed habitats in north-eastern Borneo. We measured interference competition

abilities of A. gracilipes by performing arena bioassays between two A. gracilipes colonies

and seven local ant species, and measured its effective resource competition at baits within

supercolonies and at supercolony boundaries. Furthermore, we compared ant species diversity

and composition at baits among (A) core areas of A. gracilipes supercolonies, (B)

supercolony boundaries and (C) outside supercolonies.

A. gracilipes was behaviourally dominant over most ant species except Oecophylla

smaragdina. Within supercolonies, A. gracilipes discovered all food baits first, and

monopolized the vast majority throughout the course of the experiment. At supercolony

boundaries, A. gracilipes discovered baits later than resident ant species, but subsequently

monopolized half of the baits. Furthermore, the activity and diversity of the ant community

within A. gracilipes supercolonies was lower than at its boundaries and outside supercolonies,

and the ant communities differed significantly between infested and non-infested areas. Our

study supports the hypothesis that successful establishment of A. gracilipes in

anthropogenically disturbed habitats may negatively affect resident ant communities through


74

high levels of direct interspecific aggression and almost complete monopolization of

resources within high-density supercolonies.

INTRODUCTION

Biological invasions rank among the most important drivers of environmental degradation

(Mack et al. 2000b). Successful invasions by introduced ant species have been shown to cause

decreases in diversity and abundance of both vertebrate and invertebrate populations (Suarez

et al. 2005; Wetterer and Moore 2005; Davis et al. 2008) and may even lead to alteration of

whole ecosystems, e.g. through predation on key-stone species (O'Dowd et al. 2003),

competitive advantages over native invertebrates (McNatty et al. 2009), disruption of

essential mutualisms such as seed dispersal (Christian 2001; Davis et al. 2010), pollination

(Blancafort and Gomez 2005) or disturbance of native ant communities (Holway 1999).

Ants within a community compete with each other either directly (interference competition,

predation) or indirectly (via exploitation of mutually required space or food resources).

Invasive ant species often share characteristics that render them superior in both interference

and exploitation competition (reviewed by Holway et al. 2002). Among these features are (1)

pronounced interspecific aggression (Holway 1999; Human and Gordon 1999), (2) superiority

at resource discovery and monopolization (Davidson 1998; Holway 1999) and (3) numerical

dominance within the invaded habitat (Walters and Mackay 2005). Introduced ant species

often form large polygynous and polydomous supercolonies in which large numbers of

workers move freely among nests (Buczkowski et al. 2004; Espadaler et al. 2004; Le Breton

et al. 2004; Abbott 2005). The continuous epigaeic activity of worker presence may allow

introduced ant species to outperform other ant species locally in terms of resource discovery

and monopolization (Holway 1999), as they are likely to discover resources before other ants

do in areas of high activity and have the worker capacity to monopolize it. Thus, the size of an

ant colony (in terms of worker quantity) and its competitive abilities in resource discovery

and monopolization are often linked, posing a ‘chicken and egg’ dilemma (Oliveras et al.

2005; Oliver et al. 2008). Likewise, the worker density in the foraging territory of an ant

colony and its ability to prevail in aggressive inter- and intraspecific encounters may be

intertwined (e.g. Palmer 2004). Thus, it may be important to compare outcomes of direct

competitive interactions among workers of different species in different ratios, and compare


75

resource discovery and monopolization patterns between areas of high worker density and

low worker density (e.g. in the centre of a colony and at its boundaries).

Most of our knowledge on invasive ants is derived from studies on two focal species, the

Argentine Ant Linepithema humile and the Red Imported Fire Ant (RIFA) Solenopsis invicta.

To elucidate whether the mechanisms at work in invasions by the Argentine Ant or RIFA are

of the same importance in other invasive ant species, we studied behavioural and ecological

characteristics of a potentially invasive population of supercolonies the Yellow Crazy Ant

Anoplolepis gracilipes in NE-Borneo. Despite its current distribution in tropical regions of SE

Asia and the Indopacific is well mapped, its origin (and thus its status as invasive/native in

several regions including SE-Asia) remains unclear (Wetterer 2005). Several introduced

populations have been studied on Hawaii (Kirschenbaum and Grace 2007), the Seychelles

(Gerlach 2004), Tokelau (Lester and Tavite 2004; Abbott et al. 2007) or Christmas Island

(O'Dowd et al. 2003; Abbott 2004; Thomas et al. 2010b), and potentially introduced

populations were found in Central Sulawesi (Bos et al. 2008) and NE Borneo (Drescher et al.

2007; Pfeiffer et al. 2008; Brühl and Eltz 2009). Few studies, however, aimed to identify the

mechanisms facilitating the ecological success of A. gracilipes. Those that did were limited to

Pacific islands (e.g. Lester and Tavite 2004; Abbott et al. 2007), on which native ant

communities are largely absent, or focused on the interaction of A. gracilipes with other

invading ant species (Lester and Tavite 2004; Kirschenbaum and Grace 2008).

In this study, we describe the ecological dominance of A. gracilipes supercolonies within a

resident ant community in anthropogenically altered habitats in north-eastern Borneo and

explore the underlying mechanisms by testing interference and exploitative abilities of A.

gracilipes. We measured: (1) the aggression of two different A. gracilipes laboratory colonies

in bioassays against seven resident ant species (six native species and one introduced species);

(2) the resource discovery and monopolization patterns of ants in the field by installing

artificial baits (honey, tuna) within, outside and across A. gracilipes supercolony boundaries;

and (3) the ant community composition along bait transects within, outside and across

boundaries of A. gracilipes supercolonies.


76

METHODS

All experiments were performed between October 2006 and January 2007 in Poring Hot

Springs, Sabah, Malaysia, except for bioassays using colony D1 (see below) which were

performed in March 2005 in Danum Valley, Sabah. The area in and around Poring Hot

Springs is comprised of recreational parkways (referred to as park, consisting mostly of lawn,

shrubs and bamboo) and private orchards (e.g. durian, mango, banana, rambutan) along the

edge of mildly disturbed submontane rainforest dominated by dipterocarps. Locations and

labels of Anoplolepis gracilipes supercolonies overlap with those of a previous study focusing

on population genetics (Drescher et al. 2007).

A. gracilipes was found in almost all anthropogenically altered habitats (e.g. residential area,

park, orchard and secondary forest edge) except within secondary forest. However, A.

gracilipes scouts of supercolonies located at the forest edge were observed to venture as far as

ca. 40 m into the forest. Consequently, we categorized four types of habitats, in which

experiments were performed: (1) secondary forest, (2) secondary forest edge, (3)

plantation/orchard, (4) park. Core regions of supercolonies were characterized by a

continuous above-ground activity of A. gracilipes workers foraging, carrying brood, males,

callow workers and food items from one nest or bivouac to the other. The density of A.

gracilipes workers decreased from the core region of supercolonies to its boundaries, where

workers were occasionally present as scouts or small groups of workers carrying food items.

We thus categorized three different presence/absence categories: (A) within supercolonies,

(B) at supercolony boundaries and (C) outside supercolonies. A. gracilipes was sighted in

categories (A) and (B), but not in (C).

Since our study was located in anthropogenically altered habitats, in which at least one other

introduced ant species, Tapinoma melanocephalum, was also found, we refer to the local ant

community as ‘resident’ as opposed to ‘native’.

Interspecific Aggression and numerical dominance

On two occasions, we collected subsets of two A. gracilipes supercolonies (colony D1,

Danum Valley; colony P3, Poring Hot Springs, see Drescher et. al. 2007) and transferred the

ants into plastic containers (P3: 28.2 liter, D1: 6.75 liter) with Fluon™-coated walls. The

laboratory colonies were given soil and leaves (colony D1) or cardboard and newspaper

(colony P3) as nesting material. Upon setup of the laboratory colonies, colony P3 contained at


77

least 2500 workers, 500 pupae/larvae and 13 adult queens. Colony D1 contained ca. 1500

workers, 300 pupae/larvae and 5 queens. The colonies were fed water, honey water and tuna

every three days.

In each region (Danum Valley, Poring Hot Springs), we selected colonies of several local ant

species (Oecophylla smaragdina Fabricius, Dolichoderus thoracicus Smith, Crematogaster

coriaria Mayr, C. inflata Smith, C. rogenhoferi Mayr, Technomyrmex sp. and Paratrechina

longicornis Latreille) which occurred close to supercolony boundaries (100 – 200m) in

similarly disturbed habitats and which could be sampled continuously without destroying the

colony. Only, O. smaragdina and D. thoracicus were found in both regions in sufficient

abundance to be used in the bioassays. Ant species within supercolonies were not selected,

because of their general low abundance and various sampling difficulties (e.g. arboreal nests

in Polyrachis spp., too small colony sizes in Diacamma rugosum, Odontomachus spp. or

Odontoponera spp., frequent colony movements in Leptogenys spp.).

Aggression tests were performed in PVC cylinders (height = 5 cm, diameter = 10 cm,

Fluon™-coated walls, referred to as ‘arena’) based on a sheet of paper which was replaced

after each trial. For each species combination, varying numbers of workers from colonies

D1/P3 (A. gracilipes) and from one of the seven local ant species were allowed to enter the

arena by crossing a bridge between their laboratory nests and the arena. We tested five

different ratios between workers of colony P3 and the resident ant species (1:10, 1:5, 5:5, 5:1

and 10:1) and three ratios (5:1, 5:5, 1:5) in tests of colony D1 against resident ant species,

plus 10:1 in tests against O. smaragdina. In uneven ratios, the more numerous species was

allowed into the arena first, whereas in the 5:5 ratio, each species was allowed into the arena

first in half of the replicates. After 2 min, individuals of the second ant species were

transferred into the arena in a similar fashion. When both ant species were in the arena, the

number of dead individuals of each species was counted in 5-min-intervals over a period of

60 min. In total, we performed N=432 replicates (12 replicates per ratio and species for each

colony). We calculated the average survival time for each species in every trial and

determined the difference in mean survival time for each ratio and species combination.

Composition of ant communities at bait transects

We installed three types of transects of varying length (90m-120m, see Tables 10-13) in each

of six supercolonies (resulting in N=18 transects): (A) within supercolonies, (B) across


78

supercolony boundaries and (C) outside supercolonies. Transects across supercolony

boundaries (B) were installed perpendicular to the supercolony boundary, with one half

situated within the supercolony and the other half outside the supercolony. However, since

supercolony boundaries of A. gracilipes are diffuse instead of sharply defined borders (Abbott

2006), boundaries were not always positioned centrally in each transect. Each transect

consisted of 12 evenly distributed measuring points in a straight line, and each measuring

point contained one honey bait and one tuna bait. Baits consisted of ca. 1 cm³ of honey or tuna

placed on a circular polyethylene disc (diameter = 10 cm, with a 1 cm deep central

depression). During the course of the experiment, honey and tuna were replenished wherever

necessary. The baits were observed for five hours (between 0900 h and 1600 h during dry

weather conditions) and the incidence of ant species was recorded on an hourly basis.

Wherever possible, ants were identified in the field to the species level or assigned

morphospecies. In cases where species/morphospecies affiliation was unclear, individuals

were photographed and identified by collecting up three specimen from the area surrounding

the bait disc, so that recruitment was disturbed to the least possible extent.

We compared the activity density, species richness and diversity of ant species other than A.

gracilipes among transects (A) within supercolonies, (B) across boundaries and (C) outside

supercolonies and the four habitat types (1 – secondary forest, 2 – secondary forest edge, 3 –

plantation/orchard, 4 – park). Activity density A of the entire ant community in a transect was

defined as 1

S

i

i

A a=

=∑ , with ai representing the number of baits at which species i occurs.

Species richness S was the sum of all species observed per transect (excluding A. gracilipes).

For species diversity, we used the Simpson’s reciprocal diversity index 2

1

1/ 1/S

i

i

D p=

= ∑ ,

based on relative activity density (pi) of each species (except A. gracilipes) at transects, with

ii

ap

A= . Furthermore, we compared the ant community composition in response to A.

gracilipes (A – C) and habitat type (1 – 4) by performing permutation tests (adonis, R-

package vegan 1.15, 10000 runs) based on relative activity densities (pi). To visualize the

differences in community composition among transects, we performed non-metric

multidimensional scaling (NMDS) based on Bray-Curtis-distances of relative activity

densities (pi) among transects.


79

Resource discovery and monopolization

To obtain a finer temporal resolution of species succession at resources than in the above ant

surveys, we observed baits for two hour periods in the field between 0900 h and 1600 h

during dry weather conditions. Baits were prepared and treated as above. A pair of one honey

and one tuna bait was placed 1 m apart (A) within supercolonies (three replicates), (B) at

supercolony boundaries (six replicates) and (C) outside supercolonies (five replicates). For

each bait type, we recorded the time until bait discovery for each ant species and counted the

foragers in five minute intervals. During the observation period, honey and tuna were

replenished if necessary.

RESULTS

Interspecific Aggression and numerical dominance

Generally, interspecific aggression was high and at least one individual was killed in each

trial. Anoplolepis gracilipes survived longer (i.e. killed more workers than was killed) than

many of the tested ant species in most ratios except O. smaragdina (Fig. 24 A, B), and the

patterns of competitive superiority were similar between colonies P3 and D1. In all nine pair-

wise species combinations, an increased proportion of A. gracilipes in the arena correlated

positively with superior competitive performance (Spearman rank correlation, all P < 0.01,

Fig. 24 A, B). In some cases, the superiority of A. gracilipes was already pronounced even

when ratios favored the other ant species (i.e. when outnumbered by 5:1 or 10:1 in

combinations against P3 — Technomyrmex sp., P3 — C. coriaria, D1 — C. inflata). In these

combinations, A. gracilipes workers killed more than one opponent (1.0±1.1, mean±SD)

while rarely being killed, resulting in a significantly longer survival.

Composition of ant communities at bait transects

In total, 46 ant species (excluding A. gracilipes) from 22 genera and four subfamilies were

found across all bait transects (see Supporting Tables 10 - 13 at the end of this Chapter). The

activity density (A) of resident ant species inside supercolonies was lower than at transects

across supercolony boundaries or transects outside supercolonies (ANOVA, F2, 15 = 22.5, P <

0.0001, Tukey’s post-hoc P < 0.05, Fig. 25 A), and the same trend was found for Simpson’s

diversity 1/D (F2, 15 = 13.6, P < 0.001, Tukey’s P < 0.01, Fig. 25 B) and species


80

Figure 24. Difference in survival time in arena bioassays between Anoplolepis gracilipes and resident ant species. The difference in survival time was calculated as ∆TS = TAnoplolepis – Tother ant (mean ± SEM) and arena bioassays were performed with (A) A. gracilipes supercolony P3, Poring Hot Springs, and (B) A.

gracilipes supercolony D1, Danum Valley. Worker ratios of A. gracilipes to the other ant species vary from 1:10 to 10:1. Asterisks above the brackets indicate a significant positive correlation between worker ratio and difference in survival (Spearman rank correlation). Asterisks above or below columns indicate significant deviations from zero (one-sample t-test): * p < 0.05, ** p < 0.01, *** p < 0.001.

richness S (F2, 15 = 27.0, P < 0.0001, Tukey’s P < 0.001). Neither the activity density A, ant


81

Figure 25. (A) Activity density A (number of sightings at baits) and (B) reciprocal Simpson diversity (1/D) of the resident ant community. Values are shown for transects within A. gracilipes supercolonies (IN), across supercolony boundaries (MID) and outside supercolonies (OUT) (mean ± SD). Letters indicate significant differences among groups (ANOVA, Tukey-post-hoc).

species diversity 1/D, nor richness S differed significantly among the four habitat types (1 –

secondary forest, 2 – secondary forest edge, 3 – plantation/orchard, 4 – park) (F3, 14 ≤ 2.7, P ≥

0.19) across all sample areas.

In transects across supercolony boundaries, we observed a gradual increase of A. gracilipes

worker abundance at baits when moving towards the supercolony core region (Fig. 26). The

increase of A. gracilipes forager abundance from the margins of a supercolony to its center

corresponded to a decrease in the species richness of resident ants (baits outside supercolony

boundaries were excluded from the analysis, Spearman rank correlation, R = -0.79, P = 0.033,

Fig. 26). Anoplolepis gracilipes often monopolized baits and was only observed at baits

together with other species in low numbers (e.g. when its scouts and those of other species

discovered the bait at the same time in early phases of recruitment, or when A. gracilipes

scouts discovered baits that were already being exploited by other ant species).

Ant community composition significantly differed among areas of A. gracilipes

presence/absence (adonis, F = 1.6, P = 0.019), whereas the four habitat categories did not

differ significantly (F = 0.89, P = 0.67). In the NMDS ordination (Fig. 27), this pattern is

reflected by the fact that ant assemblages cluster in response to the presence/absence of A.

gracilipes (categories A-C), but not in response to habitat categories.


82

Fig. 26. Abundance of Anoplolepis gracilipes (black squares) and number of resident ant species (white squares) at bait transects (mean ± SEM). Transects comprised 12 measuring points (10 m apart from each other), but were readjusted (post hoc) to superpose the supercolony boundaries for this analysis.

Figure 27. Multidimensional scaling (NMDS) plot of the ant assemblages at bait transects. Transects were within supercolonies (IN), across supercolony boundaries (MID) and outside Anoplolepis gracilipes supercolonies (OUT, and across four habitat types (park, orchard, forest edge, secondary forest) (Stress = 19.98, non-metric fit R² = 0.96).

Resource discovery and monopolization

Inside supercolonies, A. gracilipes was first to discover all baits, and did so within the first

minute (50.6 ± 30.6 sec, mean ± SD). Both types of baits were exploited to a similar extent

(average number of individuals at bait within two hours: 50.4 ± 14.9 for honey, 48.5 ± 12.2

for tuna). In only one case, Acanthomyrmex sp2 at a tuna bait, was any other ant species

observed to feed on any of the baits (Fig. 5 A, B). Occasionally, other ant species were

observed to approach the baits, but were aggressively chased away by A. gracilipes. At

supercolony boundaries, resident ant species discovered both honey and tuna baits

significantly earlier than A. gracilipes (resident ants: 2.4 ± 1.8 min for honey and 5.6 ± 3.1min

for tuna baits; A. gracilipes: 26.5 ± 9.1 min for honey and 16.6 ± 3.1 min for tuna baits, t-test,

all P < 0.01). Despite discovering honey baits later than resident ants, A. gracilipes

subsequently monopolized five out of six honey baits (Fig. 5 C). Tuna baits, however, were

largely monopolized by resident ants (Fig. 5 D). Outside supercolonies, the baits were fully

discovered and recruited to by the resident ant species.


83

Figure 28. Relative abundances of Anoplolepis gracilipes and resident ants at baits. Anoplolepis

gracilipes (black circles) and members of the resident ant community (white circles, pooled data) were observed on two types of baits in two types of colonization states within 120 min (mean ± SD). In Fig 5 B, Acanthomyrmex sp 2 is the only ant species other than A. gracilipes.

DISCUSSION

Anoplolepis gracilipes was highly aggressive towards other ant species and readily engaged in

lethal fights. Its workers survived much longer than workers of most other species – when a

single A. gracilipes worker was confronted with up to ten competitors, it sometimes survived

until the end of the trial and killed some competitors. Furthermore, A. gracilipes’ competitive

performance increased as its ratio in aggression tests increased. Oecophylla smaragdina was

the only ant species that matched the competitive superiority of A. gracilipes, which may in

part be due to its larger size compared to the other tested ant species including A. gracilipes,

but may also reflect its increased aggressive behaviour as a dominant ant species (Basu 1997).

This, however, would imply that the other tested ant species are less dominant, which is not

the case, e.g. in Dolichoderus thoracicus.

Under natural circumstances, ants adjust their behaviour towards potential opponents

depending on the context such as next proximity or group size (e.g. Buczkowski and


84

Silverman 2005; Sagata and Lester 2009), and may often avoid a confrontation (Human and

Gordon 1999). In arena experiments, individuals are largely deprived of the social context as

well as the chance to escape confrontation, and may thus display behaviour which differs

from natural conditions. However, arena experiments demonstrate the competitive potential of

each participant in the field, which is what was aimed for in this particular setup. The results

characterize A. gracilipes as a highly aggressive ant which may dominate other ant species in

both superior and inferior numbers, which may point towards highly efficient fighting

capabilities. In fact, within their colony territories, A. gracilipes workers were repeatedly

observed to attack and pursue foragers of other species such as Odontoponera sp. or

Diacamma sp., and dead individuals of those two and other species were frequently found

within nests of A. gracilipes (JD, pers. obs.). None of the seven ant species used for bioassays

were observed to nest within supercolony territories. This pattern may be the result of the

unilateral competitive superiority of A. gracilipes towards many ant species, while it may

reflect reciprocal competitive exclusion in the case of the likewise dominant ant species O.

smaragdina (e.g. Basu 1997; Bluthgen and Fiedler 2004).

Within supercolonies, A. gracilipes outperformed most other ant species in terms of resource

discovery and monopolization. This ability to dominate resource acquisition within territorial

boundaries is known from other invasive ant species such as Linepithema humile (Human and

Gordon 1999) and Solenopsis invicta (Calcaterra et al. 2008) and may be the cause or

consequence of local displacement of competing ant species in invaded habitats. At

supercolony boundaries, the abundance of A. gracilipes workers was lower than inside core

areas. Here, A. gracilipes discovered resources later than other ant species, but displaced

many of them after discovery. This, in turn, may give other ant species time to exploit a

resource before recruitment of A. gracilipes starts (dominance-discovery trade-off, see Fellers

1987; Davidson 1998; Holway 1999) and might facilitate species coexistence with A.

gracilipes in areas of low abundance. This status quo, however, might only be temporary, as

boundaries of A. gracilipes supercolonies are highly dynamic and may spread as much as 0.5

m per day (Abbott 2006). Furthermore, our study showed that the activity, diversity and

composition of ant communities at bait transects were affected by the presence/absence of A.

gracilipes and not by differences between the four disturbed habitat types, suggesting that A.

gracilipes may also dominate natural food resources within the study area. As competitive

exclusion from food resources by dominant ants may control ant species richness at the

assemblage level (Parr 2008), the numerical dominance of A. gracilipes at baits reliably


85

reflects its ecological dominance within the resident ant community. Acanthomyrmex sp2 was

one of the few ant morphospecies that were observed at baits within supercolonies, and the

only one that successfully monopolized a bait within A. gracilipes territory. As this species

could not be assigned to any presently known Acanthomyrmex species from Borneo, it is

unclear whether Acanthomyrmex sp2 represents a species that has the potential to may

commonly prevail in the presence of A. gracilipes or whether that observation was an

exception to the rule.

We conclude that within disturbed habitats, A. gracilipes has the potential to displace many

species of local ant communities. Its spread is facilitated by a combination of direct

interspecific aggression, resource monopolization and numerical dominance. A. gracilipes has

strong individual competitive abilities, which may aid in prevailing in new habitats as a

propagule. This study once again emphasizes the capability of invasive ants to influence

entire ant communities by dominating both interspecific encounters and exploitation of food

resources.

ACKNOWLEDGMENTS


Prof. Dr. Datin Maryati Mohamed and Dr. Arthur Chung for collaboration and support.

Albinus Ongkudon (DVMC) and Dr. Maklarin bin Laklim (Sabah Parks) facilitated our work

at Danum Valley and Poring Hot Springs. Janina Eulenburg is thanked for help with

fieldwork and data collection. This work was part of the Royal Society’s Southeast Asia

Rainforest Research Programme (SEARRP) and was funded by the Deutsche


project E2). All experiments conducted comply with the current laws of Malaysia.


86

SUPPORTING TABLES Table 10. Species composition of six transects inside Anoplolepis gracilipes supercolonies (IN) within five hours. Numbers indicate at how many of 12 measuring points a species was found. Each measuring point consisted of one honey and one tuna bait.

Subfamily Genus / species IN 1 IN 2 IN 3 IN 4 IN 5 IN 6 Formicinae Anoplolepis gracilipes 12 12 12 9 12 12 Formicinae Euprenolepis sp1 1 Paratrechina sp1 1 Paratrechina sp3 1 Paratrechina sp5 2 Paratrechina sp6 4 Paratrechina sp7 1 4 Myrmicinae Acanthomyrmex sp4 4 Cardiocondyla sp1 2 Crematogaster sp3 2 Lophomyrmex sp1 1 Lordomyrma sp1 1 Pheidologeton sp2 1 1 Ponerinae Odontomachus sp1 1 1 1 Odontoponera sp2 2 Transect length 90 120 100 90 120 90

Habitat type

park park park park orchard forest edge


87

Table 11. Species composition of six transects across Anoplolepis gracilipes supercolony boundaries (EDGE) within five hours. Numbers indicate at how many of 12 measuring points a species was found. Each measuring point consisted of one honey and one tuna bait. Subfamily Genus / species EDGE 1 EDGE 2 EDGE 3 EDGE 4 EDGE 5 EDGE 6

Formicinae Anoplolepis gracilipes 5 4 6 6 5 7 Dolichoderinae Dolichoderus sp1 1 Formicinae Camponotus sp1 1 Camponotus sp2 1 1 Camponotus sp3 1 Euprenolepis sp1 1 Paratrechina sp1 1 2 2 5 Paratrechina sp2 3 4 Paratrechina sp3 1 2 Paratrechina sp4 2 2 Polyrhachis sp2 1 Myrmicinae Acanthomyrmex sp2 3 3 Acanthomyrmex sp3 1 Acanthomyrmex sp4 3 8 4 10 Cardiocondyla sp1 1 2 Crematogaster sp2 4 Crematogaster sp3 2 5 Lordomyrma sp1 1 Monomorium sp1 3 1 3 Pheidologeton sp1 1 1 2 Pheidologeton sp2 1 2 Tetramorium sp1 2 1 Tetramorium sp2 1 Ponerinae Diacamma rugosum 5 3 3 4 3 3 Leptogenys sp1 1 Odontomachus sp1 2 1 3 Odontomachus sp2 4 1 4 Odontoponera sp2 2 2 1 1 3 Odontoponera sp3 1 1 Transect length 120 120 120 120 120 120

Habitat type park park park orchard forest edge

forest edge


88

Table 12. Species composition of six transects outside Anoplolepis gracilipes supercolony boundaries (OUT) within five hours. Numbers indicate at how many of 12 measuring points a species was found. Each measuring point consisted of one honey and one tuna bait. Subfamily Genus / species OUT 1 OUT 2 OUT 3 OUT 4 OUT 5 OUT 6

Dolichoderinae Dolichoderus sp1 1 Tapinoma melanocephalum 4 Formicinae Camponotus gigas 3 1 Camponotus sp2 4 Camponotus sp3 1 Camponotus sp4 1 Camponotus sp5 2 Camponotus sp6 1 Oecophylla smaragdina 1 Paratrechina sp1 3 1 1 Paratrechina sp2 3 2 Paratrechina sp3 1 Paratrechina sp6 10 8 Paratrechina sp7 2 6 Polyrhachis sp1 2 3 Polyrhachis sp2 1 Myrmicinae Acanthomyrmex sp1 2 4 Acanthomyrmex sp2 3 1 Acanthomyrmex sp3 3 Acanthomyrmex sp4 7 3 5 Aphaenogaster sp1 1 Cardiocondyla sp1 1 Crematogaster sp1 1 1 Crematogaster sp2 5 6 3 Crematogaster sp3 3 Lophomyrmex sp1 1 Lordomyrma sp1 1 6 Monomorium sp1 3 Myrmicaria sp1 5 Myrmicaria sp2 2 Pheidole sp1 1 1 2 2 Pheidologeton sp1 1 1 Pheidologeton sp2 1 Tetramorium sp1 1 Ponerinae Diacamma rugosum 3 4 1 3 2 Odontomachus sp1 2 6 1 1 Odontomachus sp2 6 Odontoponera sp1 2 Odontoponera sp2 3 9 4 Odontoponera sp3 1 6 Transect length 90 120 120 100 100 100

Habitat type

park forest edge

forest edge

secondary forest

secondary forest

secondary forest

IX. The cricket Myrmecophilus pallidithorax

89

IX. AGILITY DETERMINES THE SURVIVAL OF THE

CLEPTOPARASITIC CRICKET MYRMECOPHILUS

PALLIDITHORAX IN SUPERCOLONIES OF ANOPLOLEPIS

GRACILIPES

ABSTRACT

Myrmecophily is a taxonomically widespread phenomenon which includes mutualistic,

commensalistic and parasitic associations between ants and other organisms. Organisms

exploiting the resources offered by ant colonies, such as food and shelter, have evolved a

variety of morphological, behavioural and chemical mechanisms to circumvent the colonies’

defences. In this study, we examined the mechanisms employed by the myrmecophilous

cricket Myrmecophilus pallidithorax (Orthoptera, Myrmecophilidae) that enable it to survive

in supercolonies of the invasive Yellow Crazy Ant Anoplolepis gracilipes. By reducing M.

pallidithorax’ agility (through induced autotomy of the cricket’s hind legs), we were able to

show that the ability to avoid potentially lethal encounters with its hosts by evasive jumping

significantly increases the survival of M. pallidithorax in standardized A. gracilipes

subcolonies. Moreover, we observed significantly higher survival of handicapped M.

pallidithorax individuals in subcolonies that consisted of crickets and ants from the same A.

gracilipes supercolony than in subcolonies that consisted of ants and crickets from mutually

aggressive A. gracilipes supercolonies. The apparently colony-specific discrimination

suggests that the cuticular hydrocarbon (CHC) profile of M. pallidithorax may differ between

populations from different host supercolonies. Preliminary data show a high similarity

between CHC profiles of M. pallidithorax and A. gracilipes from the same supercolony. We

discuss our findings with respect to the potential mechanisms facilitating the successful

establishment of M. pallidithorax in supercolonies of its host.

INTRODUCTION

Social insects invest much of their time constructing shelters and gathering food, providing

them with protection for their queen(s) and brood and a stock of resources as an insurance


90

against variable conditions (Hölldobler and Wilson 1990; Blüthgen and Feldhaar 2010).

While the almost perpetual availability of food and shelter in social insect societies provides

the basis for many mutualistic associations that they engage in, it also represents valuable

resources that are exploited by a vast array of commensals and parasites (Hölldobler and

Wilson 1990). In ants, an estimated 80,000 – 100,000 morphologically distinct species of

myrmecophilous insects exist worldwide, 5000 – 20000 of which are thought to be obligate

social parasites (Elmes 1996). Once inside ant colonies, they prey upon colony members

(including eggs and larvae), parasitize food resources, scrounge food via trophallaxis or enjoy

social benefits such as protection from their own enemies (Hölldobler and Wilson 1990;

Dettner and Liepert 1994). Ant colonies are well defended and workers are usually hostile

towards intruders. Thus, in order to enter and live inside an ant colony, social parasites

evolved strategies to overcome the colony defences, e.g. mimicking the ants’ communication.

Members of ant communities recognize each other by a mix of tactile (antennation) and

chemical cues (mainly cuticular hydrocarbons, CHC’s)(Vander Meer and Morel 1998; Lahav

et al. 1999; Akino et al. 2004; Torres et al. 2007). Accordingly, many social parasites of ants

show morphological, behavioural and/or chemical adaptations that allow them to enter and

remain within their hosts colonies largely undetected, such as chemical mimicry or

camouflage (Dettner and Liepert 1994; Lenoir et al. 2001; Akino 2008).

In this study, we explore the adaptations that allow the myrmecophilous cricket

Myrmecophilus pallidithorax to enter and live inside nests of the invasive Yellow Crazy Ant

Anoplolepis gracilipes. M. pallidithorax (Chopard 1930)[ = M. mayaealberti, Hugel &

Matyot (2006) syn. nov. and M. leei, Kistner & Chong (2007), syn. nov., according to

Ingrisch (2010)] belongs to the ca. 40-60 species of myrmecophilous crickets described to

date (Kistner 1982; Kistner et al. 2007), which belong almost exclusively to the genus

Myrmecophilus (syn. Myrmecophila Latreille, 1927; exception: Camponophilus). Like all

Myrmecophilus crickets, M. pallidithorax is characterized by its small size (<5mm), aptery

and plump body form. Despite the fact that myrmecophilous crickets were studied as early as

1799 (Panzer 1799), little is known about their feeding behaviour in ant nests as well as the

mechanisms by which aggression from the host is averted. Some Myrmecophilus crickets may

intercept or even elicit trophallaxis of their host ants (Henderson and Akre 1986; Komatsu et

al. 2009), thereby acquiring regurgitated food such as honeydew or prey haemolymph.

Aggression from the respective host ant species seems to be avoided by escaping hostile


91

workers and by chemical camouflage, in which Myrmecophilus crickets acquire the host’s

species-specific CHC-profile by direct contact with the ants (Akino et al. 1996).

A previous study showed that A. gracilipes supercolonies may be highly differentiated

chemically up to the point that CHC profiles of workers from different supercolonies possess

a high proportion of qualitatively different hydrocarbons (Drescher et al. 2010). Provided that

M. pallidithorax uses chemical camouflage, the CHC profiles of M. pallidithorax specimen

from different A. gracilipes supercolonies may thus also differ to a great extent, possibly

resulting in higher aggression of ants towards crickets from different supercolonies than

towards crickets from their own supercolony.

This study focused on the relative importance of two potential adaptations that appear to

facilitate M. pallidithorax’ survival in supercolonies of its host, i.e. escape from precarious

situations by jumping and deception of the host by chemical camouflage. In particular, we

introduced handicapped crickets (unable to jump) into A. gracilipes subcolonies, expecting

them to suffer from a higher mortality than intact crickets that retained the ability to jump.

Furthermore, we introduced both handicapped and intact crickets into A. gracilipes

subcolonies consisting of ants sampled from the crickets’ ‘home’ supercolony as well as

‘foreign’ supercolonies, expecting them to experience higher aggression (and thus mortality)

in ‘foreign colony’ as opposed to ‘home colony’ assays.

METHODS


Anoplolepis gracilipes and Myrmecophilus pallidithorax were sampled from four different

supercolonies in NE-Borneo, two from Poring Hot Springs (6°04' N, 116°70' E; supercolonies

P3 and P6, see Drescher et al. 2010) and two from Sepilok Forest Reserve (5°51’ N, 117°57’

E, supercolonies S6 and S7). Pieces of bamboo and drainage pipes were placed next to nest

entrances within supercolonies and were readily accepted by A. gracilipes as alternative

nesting sites. Within several days, A. gracilipes supercolony subsets containing queens, males

and brood as well as M. pallidithorax and other ant guests (Stys et al. 2010) had moved in and

were subsequently transferred into plastic buckets (Vol. = 25 l) treated with Fluon™ to


92

prevent escape. All colonies were offered newspaper and cardboard as nesting material and

were fed water, honey and tuna or crayfish scraps ad libitum.

Manipulation of crickets and experimental setups

Like many crickets (Fleming et al. 2007), M. pallidithorax has a pair of hind legs which can

be shed off along a breakage plane at the femur-trochanter junction, a defensive mechanism

termed autotomy. ‘Handicapped’ crickets were generated by inducing autotomy, i.e. by

pinching the femurae of the hind legs with sharp forceps which caused the shedding of the

leg. We could not detect haemolymph leaking out of the wounds upon autotomy of the hind

legs. Nevertheless, ‘handicapped’ crickets were kept isolated for 24-48 h prior to the

experiments, thus minimizing the risk of open wounds evoking aggression by A. gracilipes or

introducing lethally wounded crickets into the experimental colonies, both of which would

have biased the experimental setup. To reduce potential differences in CHC profiles between

‘handicapped’ and ‘intact’ crickets, ‘intact’ crickets were also kept isolated for 24-48h prior to

the experiments.

Groups of ten crickets, both ‘intact’ and ‘handicapped’, were then introduced into A.

gracilipes laboratory subcolonies consisting of ants from the crickets’ ‘home’ supercolony or

‘foreign’ supercolony, or were kept without ants at all (see Table 1 for sample sizes). Due to

limited cricket availability, experiments were performed in each region separately (Poring and

Sepilok, respectively), i.e. no crickets from A. gracilipes supercolonies in Poring were

introduced into laboratory subcolonies consisting of ants from Sepilok and vice versa. For

each replicate, experimental A. gracilipes subcolonies consisted of 500 workers, three to five

males, one queen, 50 pupae/large larvae and some egg packages. Subcolonies were kept in

plastic containers (Vol. = 1100 ml) with Fluon™-coated walls. Cardboard and newspaper was

added as nesting material and subcolonies were fed water, honey and tuna/crayfish ad libitum.

Prior to the introduction of crickets, ants were given one hour to familiarize themselves with

the new environment. Every day for four consecutive days, subcolonies were screened for

dead M. pallidithorax individuals. For data analysis, we pooled the number of dead crickets in

‘home colony’ assays and ‘foreign colony’ assays, respectively. Overall, we obtained data

from 1370 crickets (137 trials containing ten crickets each, Tab. 14).


93


To obtain preliminary data on the similarity of M. pallidithorax CHC profiles and the CHC

profiles of their respective A. gracilipes host supercolonies, CHC’s from 10 adult M.

pallidithorax and 20 A. gracilipes workers from supercolony P3 were extracted with Hexane

and analyzed according to the protocol in Drescher at al. (2010). Due to the lack of replicates

from other supercolonies, the two resulting CHC profiles are simply presented and discussed

as preliminary evidence for the potential capability of M. pallidithorax to adjust their CHC

profile to the CHC profile of their host (Akino et al. 1996).

RESULTS

In all experimental setups, Myrmecophilus pallidithorax individuals were killed by the ants

throughout the experiment, with the highest mortality occurring within the first 24 h after

introduction of the crickets into A. gracilipes colonies (Fig. 29). After four days, handicapped

crickets had experienced significantly higher mortality (Ndead = 9.0 ± 1.2) than intact crickets

M. pallidithorax

‘home colony’ A. gracilipes supercolony

Ants from… Cricket treatment Nreplicates

P3 P3 home colony intact 5 P6 P6 home colony intact 5 P3 P6 foreign colony intact 5 P6 P3 foreign colony intact 5 P3 P3 home colony handicapped 10 P6 P6 home colony handicapped 10 P3 P6 foreign colony handicapped 10

Por

ing

Hot

Spr

ings

P6 P3 foreign colony handicapped 10

S6 S6 home colony intact 5 S7 S7 home colony intact 5 S6 S7 foreign colony intact 5 S7 S6 foreign colony intact 5 S6 S6 home colony handicapped 10 S7 S7 home colony handicapped 10 S6 S7 foreign colony handicapped 10

Sep

ilok

For

est

S7 S6 foreign colony handicapped 10 various - crickets alone handicapped 17

Table 14. Setup of experimental subcolonies containing Anoplolepis gracilipes and Myrmecophilus.

Pallidithorax specimen.


94

Fig. 1 Demise of Myrmecophilus pallidithorax individuals in Anoplolepis gracilipes subcolonies within four days (mean ± SD). Crickets were either untreated (‘intact’, circles) or had their hind legs removed by induced autotomy (‘handicapped’, squares). They were either kept without ants (grey squares), introduced into A. gracilipes subcolonies containing ants from the supercolony that the crickets themselves were sampled from (‘home colony’, white circles/squares) or into A. gracilipes subcolonies containing ants from different supercolonies (‘foreign colony’, black circles/squares).

in ant subcolonies (Ndead = 1.2 ± 0.9, p < 0.001, ANOVA, Tukey-Kramer HSD post-hoc test

for unequal sample sizes, Fig. 30 A) or handicapped crickets kept without ants (Ndead = 1.9 ±

0.8, p < 0.001). Moreover, handicapped crickets ‘foreign colony’ assays suffered from higher

mortality (Ndead = 9.6 ± 0.8) than handicapped crickets in ‘home colony’ assays (Ndead = 8.4 ±

1.2, p < 0.001, ANOVA, Tukey-Kramer HSD post-hoc for unequal N, Fig. 30 B). There was

no significant difference between mortality of intact crickets in ‘home colony’ assays, intact

crickets in ‘foreign colony’ assays and handicapped crickets kept without ants (p > 0.3 for all

three pairwise comparisons, Fig. 30 B). The CHC profiles of M. pallidithorax and A.

gracilipes sampled from the same supercolony showed considerable congruency (Fig. 31,

preliminary data). At least 11 of the peaks were found in profiles of both species.


95

Fig. 30 Mortality of M. pallidithorax in subcolonies of A. gracilipes at day four (mean ± SD). A. Mortality of intact crickets vs. handicapped crickets in A. gracilipes subcolonies (pooled data for intracolonial and allocolonial setups; ***: p<0.001). B. Mortality of intact and handicapped crickets in different subcolonies. Color coding is equivalent to the description in Fig. 1. Letters indicate significant differences (ANOVA, Tukey-Kramer HSD post-hoc test for unequal sample size).

DISCUSSION

Our data suggest that Myrmecophilus pallidithorax may possess two strategies that facilitate

its survival in supercolonies of Anoplolepis gracilipes: (1) avoidance of potentially precarious

interactions with its host by evasive jumping /escape, and (2) a CHC profile that shows

considerable congruency with the CHC profile of its host.

Autotomy is a defensive behavioural strategy employed by a wide variety of animals

(Maginnis 2006; Fleming et al. 2007) and it is commonly viewed as a mechanism to escape

predation, interspecific conflicts, infection or other injuries (Bateman and Fleming 2005). The

loss of a limb, however, carries costs such as increased susceptibility to predation, reduced

escape speed or lowered mating abilities (Bateman and Fleming 2005, 2006). It is commonly

accepted that in order for autotomy to have evolved, net benefits of this strategy must

outweigh net costs. Our assumption that M. pallidithorax uses autotomy as a strategy to

escape predation is based on the ease by which the hind legs were shed upon seizure with

forceps, the predominant shedding of the leg at the femur-trochanter junction (suggesting a


96

predefined breakage plane) and the observation that roughly 10% of M. pallidithorax

specimen caught under natural conditions were missing one or more limbs, including the hind

legs. Autotomy of the hind legs may be an especially useful strategy when associated with

ants, as ants tend to grasp competitors (other ants) and prey insects predominantly by the legs.

Our data suggest that the injuries caused by autotomy alone do not account for the high

mortality of handicapped crickets in A. gracilipes subcolonies (reaching 84-96%) as

handicapped crickets kept without ants suffered from much lower mortality (~ 19%). Dead

crickets that were removed from the subcolonies throughout the course of the experiment

were often shrivelled or even torn apart, further suggesting that they did not die due to direct

effects of the treatment but that they were preyed upon by A. gracilipes. In contrast, intact M.

pallidithorax specimen kept together with A. gracilipes suffered from much lower mortality

(12-13%), which signifies the importance of the hind legs for the avoidance of critical

encounters with the host species A. gracilipes.

Ant social parasites, including myrmecophiles, have repeatedly been shown to employ

chemical tactics that target ant nestmate recognition and alarm communication (Hölldobler

and Wilson 1990). These tactics include chemical mimesis (including chemical

phytomimesis, avoiding detection by ants by being ‘invisible’ via background-matching),

chemical mimicry and camouflage (avoiding attacks by ants by pretending to be nestmates)

and chemical propaganda (interfering with host nestmate recognition by creating alarm

responses)(Akino 2008). Myrmecophilous crickets from the genus Myrmecophilus spp. are

thought to employ chemical camouflage (Akino et al. 1996; Akino 2008; Komatsu et al.

2009), which differs from chemical mimicry in that CHC profiles are acquired by contact

with the host during infestation as opposed to the biosynthesis of host species-specific CHC

profiles before infestation (Dettner and Liepert 1994; Akino 2008). This difference has

important implications as chemical mimicry would suggest a coevolutionary history of

parasite and host, which in the case of M. pallidithorax is an invasive ant species of unknown

origin (Wetterer 2005). Our data imply that M. pallidithorax’ CHC-profiles may differ

between supercolonies, as significantly less crickets died in setups where ants and crickets

were from the same supercolony than when they were from different supercolonies.

Assuming that M. pallidithorax utilized chemical mimicry to reduce detection by A.

gracilipes, the severe chemical differentiation between different A. gracilipes supercolonies

(Drescher et al. 2010) would require that M. pallidithorax populations inhabiting different A.


97

Fig. 31 CHC profiles of A. gracilipes workers and M. pallidithorax in from the same supercolony. Arrows indicate the peaks/ cuticular compounds that were found in CHC profiles of both species

gracilipes supercolonies would have to biosynthesize different cuticular compounds. This, in

turn, would suggest that coevolution between A. gracilipes and M. pallidithorax occurred on

the supercolony level rather than on the species level, i.e. different M. pallidithorax

populations from different supercolonies might have to differentiate genetically and

chemically in parallel to the supercolonies they inhabited. Chemical camouflage, on the other

hand, may represent a more parsimonious explanation for the potentially different CHC

profiles of M. pallidithorax specimen from different supercolonies, as CHC profiles could be

passively acquired rather than actively biosynthesized, thus not requiring chemical

coevolution between M. pallidithorax and A. gracilipes. Adjustment of the own CHC profile

to the profile of the host could be achieved either by direct contact with the host ants or by

contact with nesting material or the refuse pile. Given the considerable aggression by A.


98

gracilipes towards M. pallidithorax, acquisition of the supercolony odour by direct contact

with the host may seem unlikely. In turn, M. pallidithorax was repeatedly observed in piles of

shed pupal skins, which were often found in areas of lower A. gracilipes densities at the fringe

of individual nests. Thus, M. pallidithorax may likely utilize chemical mimicry to deceive its

host and may possibly acquire the supercolony-specific CHC profile from shed pupal skins.

Further evidence regarding the potential utilization of chemical mimicry by M. pallidithorax

comes from its taxonomical history. According to Ingrisch (2010), M. pallidithorax (Chopard,

1930) is the original synonym of two subsequently described species (M. mayaealberti, Hugel

& Matyot, 2006 and M. leei, Kistner & Chong, 2007), one of which has been reported from

nests of the likewise invasive ant Paratrechina longicornis (Hugel and Matyot 2006). Thus,

M. pallidithorax may potentially be able to immigrate into colonies of more than just one host

species, rendering chemical mimicry (which requires coevolution with a host species) an

unlikely strategy. However, further research is needed in order to clearly identify M.

pallidithorax’ chemical strategy. In particular, analyzing the CHC profiles of M. pallidithorax

from both A. gracilipes and P. longicornis supercolonies as well as following the CHC profile

development of M. pallidithorax specimen that were freshly introduced into colonies of either

species will surely provide further evidence concerning the question whether the crickets use

chemical mimicry or chemical camouflage to reduce aggression by its host.

Conclusion

Our data suggest that avoidance of direct interactions with its host seems to be the main

strategy facilitating survival of Myrmecophilus pallidithorax in Anoplolepis gracilipes

supercolonies. Our data further suggest that CHC profiles of M. pallidithorax may differ

between A. gracilipes supercolonies and that aggression may be reduced by a CHC profile

similar to that of the host ant. The crickets’ CHC profile, however, was insufficient to

completely avoid hostility from A. gracilipes, as nearly 90% of crickets were killed by ants

from their host colony once movement was impaired by autotomy of the hind legs.


99

ACKNOWLEDGEMENTS


Prof. Dr. Datin Maryati Mohamed and Dr. Arthur Chung for collaboration and support.

Dr. Maklarin bin Lakim (Sabah Parks) and Justinah Parantis facilitated our work at Poring

Hot Springs, Syvlia Alsistu and Eugene Tan in Sepilok. Prof. Dr. K.-Eduard Linsenmair, Dr.

Brigitte Fiala and Deborah Schweinfest are thanked for important observations and helpful

comments on the experimental design. This work was part of the Royal Society’s Southeast

Asia Rainforest Research Programme (SEARRP) and was funded by the Deutsche


project E2).


100

X. References

101

X. REFERENCES

Abbott, K. L. 2004. Alien ant invasion on Christmas Island, Indian Ocean: the role of ant-scale associations in the dynamics of supercolonies of the yellow crazy ant, Anoplolepis gracilipes. PhD Thesis, Biological Sciences, Monash University, Melbourne.227pp.

Abbott, K. L. 2005. Supercolonies of the invasive yellow crazy ant, Anoplolepis gracilipes, on an oceanic island: Forager activity patterns, density and biomass. Insectes Sociaux 52: 266–273.

Abbott, K. L. 2006. Spatial dynamics of supercolonies of the invasive yellow crazy ant, Anoplolepis gracilipes, on Christmas Island, Indian Ocean. Diversity and Distributions 12: 101-110.

Abbott, K. L., S. N. J. Greaves, P. A. Ritchie, P. J. Lester. 2007. Behaviourally and genetically distinct populations of an invasive ant provide insight into invasion history and impacts on a tropical ant community. Biological Invasions 9: 453-463.

Adams, E. S. 1991. Nest-mate recognition based on heritable odors in the termite Microcerotermes arboreus. Proceedings of the National Academy of Sciences of the

United States of America 88: 2031-2034. Adams, J. M., W. Fang, R. M. Callaway, D. Cipollini, E. Newell. 2009. A cross-continental

test of the Enemy Release Hypothesis: leaf herbivory on Acer platanoides (L.) is three times lower in North America than in its native Europe. Biological Invasions 11: 1005-1016.

Akino, T. 2008. Chemical strategies to deal with ants: a review of mimicry, camouflage, propaganda, and phytomimesis by ants (Hymenoptera: Formicidae) and other arthropods. Myrmecological News 11: 173-181.

Akino, T., R. Mochizuki, M. Morimoto, R. Yamaoka. 1996. Chemical camouflage of myrmecophilous cricket Myrmecophilus sp. to be integrated with several ant species. Japanese Journal of Applied Entomology and Zoology 40: 39-46.

Akino, T., K. Yamamura, S. Wakamura, R. Yamaoka. 2004. Direct behavioral evidence for hydrocarbons as nestmate recognition cues in Formica japonica (Hymenoptera: Formicidae). Applied Entomology and Zoology 39: 381-387.

Aloo, P. A. 2003. Biological diversity of the Yala Swamp lakes, with special emphasis on fish species composition, in relation to changes in the Lake Victoria Basin (Kenya): threats and conservation measures. Biodiversity and Conservation 12: 905-920.

Aron, S. 2001. Reproductive strategy: an essential component in the success of incipient colonies of the invasive Argentine ant. Insectes Sociaux 48: 25-27.

Atkinson, I. A. E., E. K. Cameron. 1993. Human influence on the terrestrial biota and biotic communities of New Zealand. Trends in Ecology & Evolution 8: 447-451.

Baier, R. E., R. L. Forsberg, A. E. Meyer, J. J. Zambon, B. Galil, N. Hulsmann, L. A. Drake, F. C. Dobbs. 2003. Ballast water biofilms as "benchmarks" for bioinvasion risk management. Abstracts of Papers of the American Chemical Society 226: U363-U363.

Balas, M. T., E. S. Adams. 1996. Nestmate discrimination and competition in incipient colonies of fire ants. Animal Behaviour 51: 49-59.

Basu, P. 1997. Competition hierarchy in the ground foraging ant community in a wet evergreen forest (Western Ghats, India): Role of interference behaviour. Current

Science 73: 173-179. Bateman, P. W., P. A. Fleming. 2005. Direct and indirect costs of limb autotomy in field

crickets, Gryllus bimaculatus. Animal Behaviour 69: 151-159. Bateman, P. W., P. A. Fleming. 2006. Increased susceptibility to predation for autotomized

house crickets (Acheta domestica). Ethology 112: 670-677.

X. References

102

Bernasconi, G., J. E. Strassmann. 1999. Cooperation among unrelated individuals: the ant foundress case. Trends in Ecology & Evolution 14: 477-482.

Blancafort, X., C. Gomez. 2005. Consequences of the Argentine ant, Linepithema humile (Mayr), invasion on pollination of Euphorbia characias (L.) (Euphorbiaceae). Acta

Oecologica 28: 49-55. Blight, O., M. Renucci, A. Tirard, J. Orgeas, E. Provost. 2010. A new colony structure of the

invasive Argentine ant (Linepithema humile) in Southern Europe. Biological Invasions 12: 1491-1497.

Blüthgen, N., H. Feldhaar. 2010. 'Food and shelter: how resources influence ant ecology' in 'Ant Ecology', pp. 115-136, edited by L. Lach, C. L. Parr, and K. Abbott. Oxford University Press, New York.

Bluthgen, N., K. Fiedler. 2004. Competition for composition: Lessons from nectar-feeding ant communities. Ecology 85: 1479-1485.

Bolton, B. 1995. A New General Catalogue of the Ants of the World. Harvard University Press, Cambridge.

Born, W., F. Rauschmayer, I. Brauer. 2005. Economic evaluation of biological invasions - a survey. Ecological Economics 55: 321-336.

Bos, M. M., J. M. Tylianakis, I. Steffan-Dewenter, T. Tscharntke. 2008. The invasive Yellow Crazy Ant and the decline of forest ant diversity in Indonesian cacao agroforests. Biological Invasions 10: 1399-1409.

Boudouresque, C. F., M. Verlaque. 2002. Biological pollution in the Mediterranean Sea: invasive versus introduced macrophytes. Marine Pollution Bulletin 44: 32-38.

Brandstaetter, A. S., A. Endler, C. J. Kleineidam. 2008. Nestmate recognition in ants is possible without tactile interaction. Naturwissenschaften 95: 601-608.

Brandt, M., E. van Wilgenburg, R. Sulc, K. J. Shea, N. D. Tsutsui. 2009a. The scent of supercolonies: the discovery, synthesis and behavioural verification of ant colony recognition cues. Bmc Biology 7: -.

Brandt, M., E. Van Wilgenburg, N. D. Tsutsui. 2009b. Global-scale analyses of chemical ecology and population genetics in the invasive Argentine ant. Molecular Ecology 18: 997-1005.

Brühl, C. A., T. Eltz. 2009. Fuelling the biodiversity crisis: species loss of ground-dwelling forest ants in oil palm plantations in Sabah, Malaysia (Borneo). Biodiversity and

Conservation 19: 519-529. Buczkowski, G., J. Silverman. 2005. Context-dependent nestmate discrimination and the

effect of action thresholds on exogenous cue recognition in the Argentine ant. Animal

Behaviour 69: 741-749. Buczkowski, G., E. L. Vargo, J. Silverman. 2004. The diminutive supercolony: the Argentine

ants of the southeastern United States. Molecular Ecology 13: 2235-2242. Calcaterra, L. A., J. P. Livore, A. Delgado, J. A. Briano. 2008. Ecological dominance of the

red imported fire ant, Solenopsis invicta, in its native range. Oecologia 156: 411-421. Caldera, E. J., K. G. Ross, C. J. DeHeer, D. D. Shoemaker. 2008. Putative native source of the

invasive fire ant Solenopsis invicta in the USA. Biological Invasions 10: 1457-1479. Carlton, J. T. 1996. Biological invasions and cryptogenic species. Ecology 77: 1653-1655. Carlton, J. T., M. R. Ruiz. 2005. 'Vector Science and Integrated Vector Management in

Bioinvasion Ecology: Conceptual Frameworks' in 'Invasive Alien Species - A New

Synthesis', pp. 36-58, edited by H. A. Mooney, R. N. Mack, J. A. McNeely, L. E. Neville, P. J. Schei, and J. K. Waage. Island Press, Washington, D.C.

Case, T. J. 1996. Global patterns in the establishment and distribution of exotic birds. Biological Conservation 78: 69-96.

Chopard, L. 1930. The Gryllidae of Sarawak. Sarawak Museum Journal Kuching 4: 1-42.

X. References

103

Christian, C. E. 2001. Consequences of a biological invasion reveal the importance of mutualism for plant communities. Nature 413: 635-639.

Cincotta, C. L., J. M. Adams, C. Holzapfel. 2009. Testing the enemy release hypothesis: a comparison of foliar insect herbivory of the exotic Norway maple (Acer platanoides L.) and the native sugar maple (A. saccharum L.). Biological Invasions 11: 379-388.

Clausen, R., R. York. 2008. Economic growth and marine biodiversity: Influence of human social structure on decline of marine trophic levels. Conservation Biology 22: 458-466.

Clement, M., D. Posada, K. A. Crandall. 2000a. TCS: a computer program to estimate gene genealogies. Molecular Ecology 9: 1657-1659.

Clement, M., D. Posada, K. A. Crandall. 2000b. TCS: A computer program to estimate gene genealogies. Molecular Ecology 9: 1657-1659.

Colautti, R. I., A. Ricciardi, I. A. Grigorovich, H. J. MacIsaac. 2004. Is invasion success explained by the enemy release hypothesis? Ecology Letters 7: 721-733.

Corin, S. E., P. J. Lester, K. L. Abbott, P. A. Ritchie. 2007. Inferring historical introduction pathways with mitochondrial DNA: the case of introduced Argentine ants (Linepithema humile) into New Zealand. Diversity and Distributions 13: 510-518.

Crawley, M. J., ed. 1997. Plant Ecology. University Press, Cambridge. Cremer, S., L. V. Ugelvig, F. P. Drijfhout, B. C. Schlick-Steiner, F. M. Steiner, B. Seifert, D.

P. Hughes, A. Schulz, K. S. Petersen, H. Konrad, C. Stauffer, K. Kiran, X. Espadaler, P. d'Ettorre, N. Aktac, J. Eilenberg, G. R. Jones, D. R. Nash, J. S. Pedersen, J. J. Boomsma. 2008. The evolution of invasiveness in Garden Ants. Plos One 3: -.

Crooks, J. A. 2002. Characterizing ecosystem-level consequences of biological invasions: the role of ecosystem engineers. Oikos 97: 153-166.

Crozier, R. H., Y. C. Crozier. 1993. The mitochondrial genome of the honeybee Apis

mellifera: Complete sequence and genome organization. Genetics 133: 97-117. Crozier, Y. C., S. Koulianos, R. H. Crozier. 1991. An improved test for africanized honeybee

mitochondrial-DNA. Experientia 47: 968-969. Cuvillier-Hot, V., R. Gadagkar, C. Peeters, M. Cobb. 2002. Regulation of reproduction in a

queenless ant: aggression, pheromones and reduction in conflict. Proceedings of the

Royal Society of London Series B-Biological Sciences 269: 1295-1300. Cuvillier-Hot, V., A. Lenoir, R. Crewe, C. Malosse, C. Peeters. 2004. Fertility signalling and

reproductive skew in queenless ants. Animal Behaviour 68: 1209-1219. Dapporto, L., A. E. Liebert, P. T. Starks, S. Turillazzi. 2009. The relationships between

cuticular hydrocarbon composition, faunal assemblages, inter-island distance, and population genetic variation in Tuscan Archipelago wasps. Biochemical Systematics

and Ecology 37: 341-348. Davidson, D. W. 1998. Resource discovery versus resource domination in ants: a functional

mechanism for breaking the trade-off. Ecological Entomology 23: 484-490. Davis, N. E., D. J. O'Dowd, P. T. Green, R. Mac Nally. 2008. Effects of an Alien Ant

Invasion on Abundance, Behavior, and Reproductive Success of Endemic Island Birds. Conservation Biology 22: 1165-1176.

Davis, N. E., D. J. O'Dowd, R. Mac Nally, P. T. Green. 2010. Invasive ants disrupt frugivory by endemic island birds. Biology Letters 6: 85-88.

Dettner, K., C. Liepert. 1994. 'Chemical mimicry and camouflage' in 'Annual Review of

Entomology', pp. 129-154, edited by T. E. Mittler, F. J. Radovsky, and V. H. Resh. Annual Reviews Inc., Palo Alto, USA.

Drake, J. M., D. M. Lodge. 2004. Global hot spots of biological invasions: evaluating options for ballast-water management. Proceedings of the Royal Society of London Series B-

Biological Sciences 271: 575-580.

X. References

104

Drake, L. A., M. A. Doblin, F. C. Dobbs. 2007. Potential microbial bioinvasions via ships' ballast water, sediment, and biofilm. Marine Pollution Bulletin 55: 333-341.

Drescher, J., N. Blüthgen, H. Feldhaar. 2007. Population structure and intraspecific aggression in the invasive ant species Anoplolepis gracilipes in Malaysian Borneo. Molecular Ecology 16: 1453-1465.

Drescher, J., N. Blüthgen, T. Schmitt, J. Bühler, H. Feldhaar. 2010. Societies drifting apart? Behavioural, genetic and chemical differentiation between supercolonies in the Yellow Crazy Ant Anoplolepis gracilipes. Plos One 5: e13581.

Drescher, J., H. Feldhaar, N. Blüthgen. 2011. Interspecific aggression and resource monopolization of the invasive ant Anoplolepis gracilipes in Malaysian Borneo. Biotropica 43: 93-99.

Dronnet, S., C. Lohou, J. P. Christides, A. G. Bagneres. 2006. Cuticular hydrocarbon composition reflects genetic relationship among colonies of the introduced termite Reticulitermes santonensis Feytaud. Journal of Chemical Ecology 32: 1027-1042.

Durka, W., O. Bossdorf, D. Prati, H. Auge. 2005. Molecular evidence for multiple introductions of garlic mustard (Alliaria petiolata, Brassicaceae) to North America. Molecular Ecology 14: 1697-1706.

Ehrenfeld, J. G. 2010. Ecosystem Consequences of Biological Invasions. Annual Review of

Ecology, Evolution, and Systematics 41: 59-80. Eldredge, L. G., S. E. Miller. 1997. Numbers of Hawaiian Species: Supplement2, including a

review of freshwater invertebrates. Musseum Occasional Papers 48: 3-22. Elmes, G. W. 1996. 'Biological diversity of ants and their role in ecosystem function' in

'Biodiversity research and its perspectives in East Asia', pp. 33-48, edited by B. H. Lee, T. H. Kim, and B. Y. Sun. Chonbuk National University, Korea.

Elmes, G. W., T. Akino, J. A. Thomas, R. T. Clarke, J. J. Knapp. 2002. Interspecific differences in cuticular hydrocarbon profiles of Myrmica ants are sufficiently consistent to explain host specificity by Maculinea (large blue) butterflies. Oecologia 130: 525-535.

Errard, C., J. Delabie, H. Jourdan, A. Hefetz. 2005. Intercontinental chemical variation in the invasive ant Wasmannia auropunctata (Roger) (Hymenoptera, Formicidae): A key to the invasive success of a tramp species. Naturwissenschaften 92: 319-323.

Espadaler, X., S. Rey. 2001. Biological constraints and colony founding in the polygynous invasive ant Lasius neglectus (Hymenoptera, Formicidae). Insectes Sociaux 48: 159-164.

Espadaler, X., S. Rey, V. Bernal. 2004. Queen number in a supercolony of the invasive garden ant, Lasius neglectus. Insectes Sociaux 51: 232-238.

Excoffier, L., G. Laval, S. Schneider. 2005. Arlequin ver. 3.0: An integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1: 47-50.

Falush, D., M. Stephens, J. K. Pritchard. 2003. Inference of population structure using multilocus genotype data: Linked loci and correlated allele frequencies. Genetics 164: 1567-1587.

Falush, D., M. Stephens, J. K. Pritchard. 2007. Inference of population structure using multilocus genotype data: dominant markers and null alleles. Molecular Ecology Notes 7: 574-578.

Feare, C. 1999. Ants take over from rats on Birs Island, Seychelles. Bird Conservation

International 9: 95-96. Feldhaar, H., J. Drescher, N. Blüthgen. 2006. Characterization of microsatellite markers for

the invasive ant species Anoplolepis gracilipes. Molecular Ecology Notes 6: 912-914. Fellers, J. H. 1987. Interference and exploitation in a guild of woodland Ants. Ecology 68:

1466-1478.

X. References

105

Fleming, P. A., D. Muller, P. W. Bateman. 2007. Leave it all behind: a taxonomic perspective of autotomy in invertebrates. Biological Reviews (Cambridge) 82: 481-510.

Florane, C. B., J. M. Bland, C. Husseneder, A. K. Raina. 2004. Diet-mediated inter-colonial aggression in the Formosan subterranean termite Coptotermes formosanus. Journal of

Chemical Ecology 30: 2559-2574. Foucaud, J., J. Orivel, D. Fournier, J. H. C. Delabie, A. Loiseau, J. Le Breton, P. Cerdan, A.

Estoup. 2009. Reproductive system, social organization, human disturbance and ecological dominance in native populations of the little fire ant, Wasmannia

auropunctata. Molecular Ecology 18: 5059-5073. Foucaud, J., J. Orivel, A. Loiseau, J. H. C. Delabie, H. Jourdan, D. Konghouleux, M.

Vonshak, M. Tindo, J. L. Mercier, D. Fresneau, J. B. Mikissa, T. McGlynn, A. S. Mikheyev, J. Oettler, A. Estoup. 2010. Worldwide invasion by the little fire ant: routes of introduction and eco-evolutionary pathways. Evolutionary Applications 3: 363-374.

Fournier, D., J. C. Biseau, S. Aron. 2009. Genetics, behaviour and chemical recognition of the invading ant Pheidole megacephala. Molecular Ecology 18: 186-199.

Fournier, D., A. Estoup, R. M. Orivel, J. Foucaud, H. Jourdan, J. Le Breton, L. Keller. 2005. Clonal reproduction by males and females in the little fire ant. Nature 435: 1230-1234.

Franks, N., M. Blum, R. K. Smith, A. B. Allies. 1990. Behavior and chemical disguise of cuckoo ant Leptothorax kutteri in relation to its host Leptothorax acervorum. Journal

of Chemical Ecology 16: 1431-1444. Geller, J. B., J. A. Darling, J. T. Carlton. 2010. Genetic Perspectives on Marine Biological

Invasions. Annual Review of Marine Science 2: 367-393. Gerlach, J. 2004. Impact of the invasive crazy ant Anoplolepis gracilipes on Bird Island,

Seychelles. Journal of Insect Conservation 8: 15–25. Giraud, T., J. S. Pedersen, L. Keller. 2002. Evolution of supercolonies: The Argentine ants of

southern Europe. Proceedings of the National Academy of Sciences of the United

States of America 99: 6075-6079. Glancey, B. M., J. C. E. Nickerson, D. Wojcik, J. Trager, W. A. Banks, C. T. Adams. 1987.

The increasing incidence of the polygynous form of the red imported fire ant, Solenopsis invicta (Hymenoptera, Formicidae), in Florida. Florida Entomologist 70: 400-402.

Grant, W. S., B. W. Bowen. 1998. Shallow population histories in deep evolutionary lineages of marine fishes: insights from sardines and anchovies and lessons for conservation. Journal of Heredity 89: 415-426.

Grapputo, A., S. Boman, L. Lindstrom, A. Lyytinen, J. Mappes. 2005. The voyage of an invasive species across continents: genetic diversity of North American and European Colorado potato beetle populations. Molecular Ecology 14: 4207-4219.

Hall, T. A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acid Symposium Series 41: 95-98.

Harris, R., K. Abbott, K. Barton, J. Berry, W. Don, D. Gunawardana, P. Lester, J. Rees, M. Stanley, A. Sutherland, R. Toft. 2005. Invasive ant pest risk assessment project for Biosecurity New Zealand. Series of unpublished Landcare Research contract reports to Biosecurity New Zealand. (BAH/35/2004-1.

Hefetz, A. 2007. The evolution of hydrocarbon pheromone parsimony in ants (Hymenoptera: Formicidae) - interplay of colony odor uniformity and odor idiosynchrasy. A review. Myrmecological News 10: 59-68.

Helanterä, H., J. E. Strassmann, J. Carrillo, D. C. Queller. 2009. Unicolonial ants: where do they come from, what are they and where are they going? Trends in Ecology &

Evolution 24: 341-349.

X. References

106

Henderson, G., R. D. Akre. 1986. Biology of the myrmecophilous cricket, Myrmecophila

manni (Orthoptera: Gryllidae). Journal of the Kansas Entomological Society 59: 454-467.

Hewitt, C. L., M. L. Campbell, R. E. Thresher, R. B. Martin, S. Boyd, B. F. Cohen, D. R. Currie, M. F. Gomon, M. J. Keough, J. A. Lewis, M. M. Lockett, N. Mays, M. A. McArthur, T. D. O'Hara, G. C. B. Poore, D. J. Ross, M. J. Storey, J. E. Watson, R. S. Wilson. 2004. Introduced and cryptogenic species in Port Phillip Bay, Victoria, Australia. Marine Biology 144: 183-202.

Hill, M., K. Holm, T. Vel, N. J. Shah, P. Matyot. 2003. Impact of the introduced yellow crazy ant Anoplolepis gracilipes on Bird Island, Seychelles. Biodiversity and Conservation 12: 1969–1984.

Hoffmann, B. D., C. L. Parr. 2008. An invasion revisited: the African big-headed ant (Pheidole megacephala) in northern Australia. Biological Invasions 10: 1171-1181.

Hoffmann, B. D., W. C. Saul. 2010. Yellow crazy ant (Anoplolepis gracilipes) invasions within undisturbed mainland Australian habitats: no support for biotic resistance hypothesis. Biological Invasions 12: 3093-3108.

Hölldobler, B., E. O. Wilson. 1977. The number of queens: An important trait in ant evolution. Naturwissenschaften 64: 8-15.

Hölldobler, B., E. O. Wilson. 1990. The Ants. The Belknap Press of Harvard University, Cambridge, Massachusetts.

Hölldobler, B., E. O. Wilson. 2009. The superorganism: the beauty, elegance, and

strangeness of insect societies. W. W. Norton & Company, Inc., New York. Holway, D. A. 1998. Effect of Argentina ant invasions on ground-dwelling arthropods in

northern California riparian woodlands. Oecologia 116: 252-258. Holway, D. A. 1999. Competitive mechanisms underlying the displacement of native ants by

the invasive Argentine ant. Ecology 80: 238-251. Holway, D. A., L. Lach, A. V. Suarez, N. D. Tsutsui, T. J. Case. 2002. The causes and

consequences of ant invasions. Annual Reviews of Ecology and Systematics 33: 181-233.

Holway, D. A., A. V. Suarez, T. J. Case. 1998. Loss of intraspecific aggression in the success of a widespread invasive social insect. Science 282: 949-952.

Hugel, S., P. Matyot. 2006. A new Myrmecophilus from Seychelles (Orthoptera, Grylloidea). Bulletin de la Societe Entomologique de France 111: 469-473.

Human, K. G., D. M. Gordon. 1999. Behavioral interactions of the invasive Argentine ant with native ant species. Insectes Sociaux 46: 159-163.

Ignacio, B. L., L. M. Julio, A. O. R. Junqueira, M. A. G. Ferreira-Silva. 2010. Bioinvasion in a Brazilian Bay: Filling Gaps in the Knowledge of Southwestern Atlantic Biota. Plos

One 5: e13065. Ingrisch, S. 2010. Review of the ant loving crickets from South-East Asia (Orthoptera:

Gryllidae, Myrmecophilinae). Entomologische Zeitschrift 120: 3-12. Iriarte, P. J. F., J. Balanya, M. Pascual, F. Mestres, E. R. Hasson, A. Fontdevila, L. Serra.

2009. Tracking the origin of an invasive species: Drosophila subobscura in Argentina. Journal of Evolutionary Biology 22: 650-658.

Ithaka, R., R. Gentleman. 1996. R: A language for data analysis and graphics. Journal of

Computational and Graphical Statistics 5: 299-314. Ito, F., S. Higashi. 1991. A linear dominance hierarchy regulating reproduction and

polyethism of the queenless ant Pachycondyla sublaevis. Naturwissenschaften 78: 80-82.

X. References

107

Jaquiery, J., V. Vogel, L. Keller. 2005. Multilevel genetic analyses of two European supercolonies of the Argentine ant, Linepithema humile. Molecular Ecology 14: 589-598.

Jerdon, T. C. 1851. A catalogue of the species of ants found in Southern India. Madras

Journal of Literature and Science 17. Jeschke, J. M., D. L. Strayer. 2005. Invasion success of vertebrates in Europe and North

America. Proceedings of the National Academy of Sciences of the United States of

America 102: 7198-7202. Jones, G. P., M. I. McCormick, M. Srinivasan, J. V. Eagle. 2004. Coral decline threatens fish

biodiversity in marine reserves. Proceedings of the National Academy of Sciences of

the United States of America 101: 8251-8253. Jousson, O., J. Pawlowski, L. Zaninetti, F. W. Zechman, F. Dini, G. Di Guiseppe, R.

Woodfield, A. Millar, A. Meinesz. 2000. Invasive alga reaches California. Nature 408: 157-158.

Keane, R. M., M. J. Crawley. 2002. Exotic plant invasions and the enemy release hypothesis. Trends in Ecology & Evolution 17: 164-170.

Keller, L., K. G. Ross. 1999. Major gene effects on phenotype and fitness: The relative roles of Pgm-3 and Gp-9 in introduced populations of the fire ant Solenopsis invicta. Journal

of Evolutionary Biology 12: 672-680. Kenis, M., M. A. Auger-Rozenberg, A. Roques, L. Timms, C. Pere, M. Cock, J. Settele, S.

Augustin, C. Lopez-Vaamonde. 2009. Ecological effects of invasive alien insects. Biological Invasions 11: 21-45.

Kiritani, K., K. Yamamura. 2003. 'Exotic Insects and Their Pathways for Invasion' in 'Invasive Species - Vectors and Managment Strategies', pp. 44-67, edited by G. M. Ruiz, and J. T. Carlton. Island Press, Washington, D.C.

Kirschenbaum, R., J. K. Grace. 2007. Dominant ant species in four habitats in Hawaii (Hymenoptera : Formicidae). Sociobiology 50: 1069-1073.

Kirschenbaum, R., J. K. Grace. 2008. Agonistic interactions among invasive ant species (Hymenoptera : Formicidae) from two habitats on Oahu, Hawaii. Sociobiology 51: 543-553.

Kistner, D. H. 1982. 'The social insects' bestiary' in 'Social Insects vol. III', pp. 1-244, edited by H. R. Hermann. Academic Press, New York.

Kistner, D. H., K. F. Chong, C. Y. Lee. 2007. A new Malaysian myrmecophilous cricket (Orthoptera : Myrmecophilidae). Sociobiology 50: 173-182.

Klimes, P., M. Janda, S. Ibalim, J. Kua, V. Novotny. 2011. Experimental suppression of ants foraging on rainforest vegetation in New Guinea: testing methods for a whole-forest manipulation of insect communities. Ecological Entomology 36: 94-103.

Komatsu, T., M. Maruyama, T. Itino. 2009. Behavioral differences between two ant cricket species in Nansei Islands: host-specialist versus host-generalist. Insectes Sociaux 56: 389-396.

Krieger, M. J. B. 2005. To b or not to b: A pheromone-binding protein regulates colony social organization in fire ants. Bioessays 27: 91-99.

Krushelnycky, P. D., L. L. Loope, N. J. Reimer. 2005. The Ecology, Policy, and Management of Ants in Hawaii. Proceedings of the Hawaiian Entomological Society 37: 1-25.

Lach, L., L. M. Hooper-Bùi. 2010. 'Consequences of Ant Invasions' in 'Ant Ecology', pp. 261-286, edited by L. Lach, C. L. Parr, and K. Abbott. Oxford University Press Inc., New York.

Lach, L., C. V. Tillberg, A. V. Suarez. 2010. Contrasting effects of an invasive ant on a native and an invasive plant. Biological Invasions 12: 3123-3133.

X. References

108

Lahav, S., V. Soroker, A. Hefetz, R. K. Vander Meer. 1999. Direct behavioral evidence for hydrocarbons as ant recognition discriminators. Naturwissenschaften 86: 246-249.

Le Breton, J., J. H. C. Delabie, J. Chazeau, A. Dejean, H. Jourdan. 2004. Experimental evidence of large-scale unicoloniality in the tramp ant Wasmannia auropunctata

(Roger). Journal of Insect Behavior 17: 263-271. Leniaud, L., A. Pichon, P. Uva, A. G. Bagneres. 2009. Unicoloniality in Reticulitermes urbis:

a novel feature in a potentially invasive termite species. Bulletin of Entomological

Research 99: 1-10. Lenoir, A., P. D'Ettorre, C. Errard, A. Hefetz. 2001. Chemical ecology and social parasitism

in ants. Annual Review of Entomology 46: 573-599. Lenoir, A., D. Fresneau, C. Errard, A. Hefetz. 1999. 'Individuality and colonial identity in

ants: the emergence of the social representation concept' in 'Information processing in

social insects', pp. 219-237, edited by C. Detrain, D. J. L., and J. M. Pasteels. Birkhäuser Verlag, Basel.

Lessard, J. P., J. A. Fordyce, N. J. Gotelli, N. J. Sanders. 2009. Invasive ants alter the phylogenetic structure of ant communities. Ecology 90: 2664-2669.

Lester, P. J., K. L. Abbott, M. Sarty, K. C. Burns. 2009. Competitive assembly of South Pacific invasive ant communities. BMC Ecology 9: Article No.: 3.

Lester, P. J., A. Tavite. 2004. Long-legged ants, Anoplolepis gracilipes (Hymenoptera: Formicidae), have invaded Tokelau, changing composition and dynamics of ant and invertebrate communities. Pacific Science 58: 391–401.

Liang, D., J. Silverman. 2000. 'You are what you eat': Diet modifies cuticular hydrocarbons and nestmate recognition in the Argentine ant, Linepithema humile. Naturwissenschaften 87: 412-416.

Librado, P., J. Rozas. 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25: 1451-1452.

Liu, H., P. Stiling. 2006. Testing the enemy release hypothesis: a review and meta-analysis. Biological Invasions 8: 1535-1545.

Liu, H., P. Stiling, R. W. Pemberton, J. Pena. 2006. Insect herbivore faunal diversity among invasive, non-invasive and native Eugenia species: Implications for the enemy release hypothesis. Florida Entomologist 89: 475-484.

Loope, L. L., D. Mueller-Dombois. 1989. 'Characteristics of invaded islands, with speciel reference to Hawaii' in 'Biological Invasions - A global perspective', pp. 257-281, edited by J. A. Drake, H. A. Mooney, F. Di Castri, K. H. Groves, F. S. Kruger, M. Rejmánek, and M. Williamson. Wiley, New York.

Lowe, S., M. Browne, S. Boudjelas, M. De Poorter. 2000. 100 of the world’s worst invasive alien species: A selection from the Global Invasive Species Database (GISD). Aliens 12: 12.

MA. 2003. Ecosystems and Human Well-being: A Framework for Assessment in I. P. Millenium Ecosystem Assessment, Washington, D. C. , ed. Conceptual Framework Working Group of the Millenium Ecosystem Assessment (MA), Island Press, Washington, D. C.

Mack, R. N., D. Simberloff, W. M. Lonsdale, H. Evans, M. Clout, F. A. Bazzaz. 2000a. Biotic invasions: Causes, epidemiology, global consequences, and control. Ecological

Applications 10: 689-710. Mack, R. N., D. Simberloff, W. M. Lonsdale, H. Evans, M. Clout, F. A. Bazzaz. 2000b.

Biotic invasions: Causes, epidemiology, global consequences, and control. Ecological

Applications 10: 689-710.

X. References

109

Macom, T. E., S. D. Porter. 1996. Comparison of polygyne and monogyne red imported fire ant (Hymenoptera: Formicidae) population densities. Annals of the Entomological

Society of America 89: 535-543. Maginnis, T. L. 2006. The costs of autotomy and regeneration in animals: a review and

framework for future research. Behavioral Ecology 17: 857-872. Martin, S., F. Drijfhout. 2009. A review of ant cuticular hydrocarbons. Journal of Chemical

Ecology 35: 1151-1161. Martin, S. J., H. Helanterä, F. P. Drijfhout. 2008a. Colony-specific hydrocarbons identify nest

mates in two species of Formica ant. Journal of Chemical Ecology 34: 1072-1080. Martin, S. J., H. Helanterä, F. P. Drijfhout. 2008b. Evolution of species-specific cuticular

hydrocarbon patterns in Formica ants. Biological Journal of the Linnean Society 95: 131-140.

Martin, S. J., H. Helanterä, K. Kiss, Y. R. Lee, F. P. Drijfhout. 2009. Polygyny reduces rather than increases nestmate discrimination cue diversity in Formica exsecta ants. Insectes

Sociaux 56: 375-383. Matsui, S., T. Kikuchi, K. Akatani, S. Horie, M. Takagi. 2009. Harmful effects of invasive

Yellow Crazy Ant Anoplolepis gracilipes on three island bird species of Minami-daito Island. Ornithological Science 8: 81-86.

Mayr, E. 1942. Systematics and the Origin of Species. Columbia University Press, New York, USA.

McDonald, I. A. W., J. Cooper. 1995. 'Insular lessons for global biodiversity conservation with particular reference to alien invasions' in 'Islands: Biological Diversity and

Ecosystem Function', pp. 189-204, edited by P. M. Vitousek, L. L. Loope, and H. Anderson. Springer Verlag, Berlin.

McGlynn, T. P. 1999a. Non-native ants are smaller than related native ants. The American

Naturalist 154: 690-699. McGlynn, T. P. 1999b. The worldwide transfer of ants: Geographical distribution and

ecological invasions. Journal of Biogeography 26: 535-548. McNatty, A., K. L. Abbott, P. J. Lester. 2009. Invasive ants compete with and modify the

trophic ecology of hermit crabs on tropical islands. Oecologia 160: 187-194. Messing, R. H., M. G. Wright. 2006. Biological control of invasive species: solution or

pollution? Frontiers in Ecology and the Environment 4: 132-140. Mikheyev, A. S., U. G. Mueller. 2007. Genetic relationships between native and introduced

populations of the little fire ant Wasmannia auropunctata. Diversity and Distributions 13: 573-579.

Mitchell, C. E., A. G. Power. 2003. Release of invasive plants from fungal and viral pathogens. Nature 421: 625-627.

Moller, H. 1996. Lessons for invasion theory from social insects. Biological Conservation 78: 125-142.

Monnin, T., C. Peeters. 1999. Dominance hierarchy and reproductive conflicts among subordinates in a monogynous queenless ant. Behavioral Ecology 10: 323-332.

Mooney, H. A. 2005. 'Invasive Alien Species: The nature of the problem' in 'Invasive Alien

Species - A New Synthesis', pp. 1-15, edited by H. A. Mooney, R. N. Mack, J. A. McNeely, L. E. Neville, P. J. Schei, and J. K. Waage. Island Press, Washington, D.C.

Mooney, H. A., E. E. Cleland. 2001. The evolutionary impact of invasive species. Proceedings of the National Academy of Sciences of the United States of America 98: 5446-5451.

Mooney, H. A., R. N. Mack, J. A. McNeely, L. E. Neville, P. J. Schei, J. K. Waage, eds. 2005. Invasive Alien Species - A New Synthesis. Island Press, Washington, D.C.

X. References

110

Moreau, C. S., C. D. Bell, R. Vila, S. B. Archibald, N. E. Pierce. 2006. Phylogeny of the ants: Diversification in the age of angiosperms. Science 312: 101-104.

Morrison, L. W. 2000. Mechanisms of interspecific competition among an invasive and two native fire ants. Oikos 90: 238-252.

Moulton, M. P., S. L. Pimm. 1986. 'Species introductions to Hawaii' in 'Ecology of biological

invasions of North America and Hawaii', pp. 231-249, edited by H. A. Mooney, and J. A. Drake. Springer, New York.

Myers, J. H., D. Simberloff, A. M. Kuris, J. R. Carey. 2000. Eradication revisited: dealing with exotic species. Trends in Ecology & Evolution 15: 316-320.

Nelson, E., G. Mendoza, J. Regetz, S. Polasky, H. Tallis, D. R. Cameron, K. M. A. Chan, G. C. Daily, J. Goldstein, P. M. Kareiva, E. Lonsdorf, R. Naidoo, T. H. Ricketts, M. R. Shaw. 2009. Modeling multiple ecosystem services, biodiversity conservation, commodity production, and tradeoffs at landscape scales. Frontiers in Ecology and the

Environment 7: 4-11. Neves, C. S., R. M. da Rocha. 2008. Introduced and cryptogenic species and their

management in Paranagua Bay, Brazil. Brazilian Archives of Biology and Technology 51: 623-633.

O'Dowd, D. J., P. T. Green, P. S. Lake. 1999. Status, impact, and recommendations for research and management of exotic invasive ants in Christmas Island National Park. Centre for Analysis and Management of Biological Invasions, Monash University, Clayton, Victoria.

O'Dowd, D. J., P. T. Green, P. S. Lake. 2003. Invasional 'meltdown' on an oceanic island. Ecology Letters 6: 812-817.

Olesen, J. M., L. I. Eskildsen, S. Venkatasamy. 2002. Invasion of pollination networks on oceanic islands: importance of invader complexes and endemic super generalists. Diversity and Distributions 8: 181-192.

Oliver, T. H., T. Pettitt, S. R. Leather, J. M. Cook. 2008. Numerical abundance of invasive ants and monopolisation of exudate-producing resources - a chicken and egg situation. Insect Conservation and Diversity 1: 208-214.

Oliveras, J., J. M. Bas, C. Gomez. 2005. Long-term consequences of the alteration of the seed dispersal process of Euphorbia characias due to the Argentine ant invasion. Ecography 28: 662-672.

Palmer, T. M. 2004. Wars of attrition: colony size determines competitive outcomes in a guild of African acacia ants. Animal Behaviour 68: 993-1004.

Panzer, G. W. F. 1799. Faunae insectorum germanicae initia. Deutschlands Insecten. Sechster Jahrgang (Hefte LXI - LXXII. Felsecker, Nürnberg.

Parr, C. L. 2008. Dominant ants can control assemblage species richness in a South African savanna. Journal of Animal Ecology 77: 1191-1198.

Passera, L., S. Aron. 1993. Social-Control over the Survival and Selection of Winged Virgin Queens in an Ant without Nuptial Flight - Iridomyrmex-Humilis. Ethology 93: 225-235.

Payne, C. M., C. V. Tillberg, A. V. Suarez. 2004. Recognition systems and biological invasions. Annales Zoologici Fennici 41: 843-858.

Pearcy, M., M. A. D. Goodisman, L. Keller. 2011. Sib mating wothout inbreeding in the longhorn crazy ant. Proceedings of the Royal Society B - Biological Sciences in press.

Pedersen, J. S., M. J. B. Krieger, V. Vogel, T. Giraud, L. Keller. 2006. Native supercolonies of unrelated individuals in the invasive Argentive Ant. Evolution 60: 782-791.

Pennanec'h, M., L. Bricard, G. Kunesch, J. M. Jallon. 1997. Incorporation of fatty acids into cuticular hydrocarbons of male and female Drosophila melanogaster. Journal of Insect

Physiology 43: 1111-1116.

X. References

111

Perdereau, E., A. G. Bagneres, S. Dupont, F. Dedeine. 2010. High occurrence of colony fusion in a European population of the American termite Reticulitermes flavipes. Insectes Sociaux 57: 393-402.

Perrings, C., H. Mooney, M. Williamson. 2010. 'The Problem of Biological Invasions' in 'Bioinvasions and Globalization', edited by C. Perrings, H. Mooney, and M. Williamson. Oxford University Press Inc., New York.

Perrings, C., M. Williamson, E. B. Barbier, D. Delfino, S. Dalmazzone, J. Shogren, P. Simmons, A. Watkinson. 2002. Biological invasion risks and the public good: an economic perspective. Conservation Ecology 6: -.

Pfeiffer, M., H. C. Tuck, T. C. Lay. 2008. Exploring arboreal ant community composition and co-occurrence patterns in plantations of oil palm Elaeis guineensis in Borneo and Peninsular Malaysia. Ecography 31: 21-32.

Pimentel, D. 1993. 'Habitat factors in new pest invasions' in 'Evolution of Insect Pests', pp. 165-181, edited by K. C. Kim, and B. A. McPheron. Wiley, New York.

Pimentel, D., R. Zuniga, D. Morrison. 2005. Update on the environmental and economic costs associated with alien-invasive species in the United States. Ecological Economics 52: 273-288.

Pritchard, J. K., M. Stephens, P. Donnelly. 2000. Inference of population structure using multilocus genotype data. Genetics 155: 945-959.

Quek, S. P., S. J. Davies, P. S. Ashton, T. Itino, N. E. Pierce. 2007. The geography of diversification in mutualistic ants: a gene's-eye view into the Neogene history of Sundaland rain forests. Molecular Ecology 16: 2045-2062.

Queller, D. C., K. F. Goodnight. 1989. Estimating relatedness using genetic markers. Evolution 43: 258-275.

Reeve, H. K. 1989. The evolution of conspecific acceptance thresholds. The American

Naturalist 133: 407-435. Rodda, G. H., J. A. Savidge. 2007. Biology and impacts of pacific island invasive species. 2.

Boiga irregularis, the Brown Tree Snake (Reptilia : Colubiridae). Pacific Science 61: 307-324.

Ruiz, G. M., J. T. Carlton, eds. 2003. Invasive Species - Vectors and Managment Strategies. Island Press, Washington, D. C.

Ruiz, G. M., T. K. Rawlings, F. C. Dobbs, L. A. Drake, T. Mullady, A. Huq, R. R. Colwell. 2000. Global spread of microorganisms by ships - Ballast water discharged from vessels harbours a cocktail of potential pathogens. Nature 408: 49-50.

Sagata, K., P. J. Lester. 2009. Behavioural plasticity associated with propagule size, resources, and the invasion success of the Argentine ant Linepithema humile. Journal

of Applied Ecology 46: 19-27. Sala, O. E., F. S. Chapin, J. J. Armesto, E. Berlow, J. Bloomfield, R. Dirzo, E. Huber-

Sanwald, L. F. Huenneke, R. B. Jackson, A. Kinzig, R. Leemans, D. M. Lodge, H. A. Mooney, M. Oesterheld, N. L. Poff, M. T. Sykes, B. H. Walker, M. Walker, D. H. Wall. 2000. Biodiversity - Global biodiversity scenarios for the year 2100. Science 287: 1770-1774.

Sanders, N. J., N. J. Gotelli, N. E. Heller, D. M. Gordon. 2003. Community disassembly by an invasive species. Proceedings of the National Academy of Sciences of the United States

of America 100: 2474-2477. Savidge, J. A. 1987. Extinction of an island forest avifauna by an introduced snake. Ecology

68: 660-668. Sax, D. F., J. H. Brown. 2000. The paradox of invasion. Global Ecology and Biogeography 9:

363-371.

X. References

112

Sodhi, N. S., L. P. Koh, B. W. Brook, P. K. L. Ng. 2004. Southeast Asian biodiversity: an impending disaster. Trends in Ecology & Evolution 19: 654-660.

Stachowicz, J. J., H. Fried, R. W. Osman, R. B. Whitlatch. 2002. Biodiversity, invasion resistance, and marine ecosystem function: Reconciling pattern and process. Ecology 83: 2575-2590.

Steiner, F. M., B. C. Schlick-Steiner, K. Moder, C. Stauffer, W. Arthofer, A. Buschinger, X. Espadaler, E. Christian, K. Einfinger, E. Lorbeer, C. Schafellner, M. Ayasse, R. H. Crozier. 2007. Abandoning aggression but maintaining self-nonself discrimination as a first stage in ant supercolony formation. Current Biology 17: 1903-1907.

Stys, P., P. Banar, J. Drescher. 2010. A new Oncylocotis Stal (Hemiptera: Heteroptera: Enicocephalidae) from Sabah: a predacious species associated with the yellow crazy ant, Anoplolepis gracilipes (Smith) (Hymenoptera: Formicidae). Zootaxa: 27-42.

Suarez, A. V., T. J. Case. 2002. Bottom-up effects on persistence of a specialist predator : Ant invasions and horned lizards. Ecological Applications 12: 291-298.

Suarez, A. V., D. A. Holway, D. S. Liang, N. D. Tsutsui, T. J. Case. 2002. Spatiotemporal patterns of intraspecific aggression in the invasive Argentine ant. Animal Behaviour 64: 697-708.

Suarez, A. V., D. A. Holway, N. D. Tsutsui. 2008. Genetics and behaviour of a colonizing species: The invasive Argentine Ant. The American Naturalist 172: 72-84.

Suarez, A. V., P. Yeh, T. J. Case. 2005. Impacts of Argentine ants on avian nesting success. Insectes Sociaux 52: 378-382.

Sunamura, E., X. Espadaler, H. Sakamoto, S. Suzuki, M. Terayama, S. Tatsuki. 2009a. Intercontinental union of Argentine ants: behavioral relationships among introduced populations in Europe, North America, and Asia. Insectes Sociaux 56: 143-147.

Sunamura, E., S. Hatsumi, S. Karino, K. Nishisue, M. Terayama, O. Kitade, S. Tatsuki. 2009b. Four mutually incompatible Argentine ant supercolonies in Japan: inferring invasion history of introduced Argentine ants from their social structure. Biological

Invasions 11: 2329-2339. Swanson, T. M. 1995. 'Why does biodiversity decline? The analysis of forces for global

change' in 'The economics and ecology of biodiversity decline - The forces driving

global change', pp. 1-10, edited by T. M. Swanson. Cambrisge University Press, Cambridge.

Takahashi, A., S. C. Tsaur, J. A. Coyne, C. I. Wu. 2001. The nucleotide changes governing cuticular hydrocarbon variation and their evolution in Drosophila melanogaster. Proceedings of the National Academy of Sciences of the United States of America 98: 3920-3925.

Thomas, M. L., K. Becker, K. Abbott, H. Feldhaar. 2010a. Supercolony mosaics: Two different invasions by the yellow crazy ant, Anoplolepis gracilipes, on Christmas Island, Indian Ocean. Biological Invasions 12: 677-687.

Thomas, M. L., K. Becker, K. L. Abbott, H. Feldhaar. 2010b. Supercolony mosaics: two different invasions by the yellow crazy ant, Anoplolepis gracilipes, on Christmas Island, Indian Ocean. Biological Invasions 12: 677-687.

Thomas, M. L., C. M. Payne-Makrisa, A. V. Suarez, N. D. Tsutsui, D. A. Holway. 2006. When supercolonies collide: territorial aggression in an invasive and unicolonial social insect. Molecular Ecology 15: 4303-4315.

Thompson, J. D., D. G. Higgins, T. J. Gibson. 1994. Clustal-W - improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Research 22: 4673-4680.

X. References

113

Tilman, D. 1997. Community invasibility, recruitment limitation, and grassland biodiversity. Ecology 78: 81-92.

Torchin, M. E., K. D. Lafferty, A. P. Dobson, V. J. McKenzie, A. M. Kuris. 2003. Introduced species and their missing parasites. Nature 421: 628-630.

Torres, C. W., M. Brandt, N. D. Tsutsui. 2007. The role of cuticular hydrocarbons as chemical cues for nestmate recognition in the invasive Argentine ant (Linepithema

humile). Insectes Sociaux 54: 363-373. Tsutsui, N. D., A. V. Suarez, D. A. Holway, T. J. Case. 2000. Reduced genetic variation and

the success of an invasive species. Proceedings of the National Academy of Sciences of

the United States of America 97: 5948-5953. Ugelvig, L. V., F. P. Drijfhout, D. J. C. Kronauer, J. J. Boomsma, J. S. Pedersen, S. Cremer.

2008. The introduction history of invasive garden ants in Europe: Integrating genetic, chemical and behavioural approaches. BMC Biology 6.

van Wilgenburg, E., C. W. Torres, N. D. Tsutsui. 2010. The global expansion of a single ant supercolony. Evolutionary Applications 3: 136-143.

van Zweden, J. S., J. B. Brask, J. H. Christensen, J. J. Boomsma, T. A. Linksvayer, P. d'Ettorre. 2010. Blending of heritable recognition cues among ant nestmates creates distinct colony gestalt odours but prevents within-colony nepotism. Journal of

Evolutionary Biology 23: 1498-1508. van Zweden, J. S., S. Dreier, P. d'Ettorre. 2009. Disentangling environmental and heritable

nestmate recognition cues in a carpenter ant. Journal of Insect Physiology 55: 158-163. Vander Meer, R. K., L. Morel. 1998. 'Nestmate recognition in ants' in 'Pheromone

communication in social insects', pp. 79-103, edited by R. K. Vander Meer, M. D. Breed, K. E. Espelie, and M. L. Winston. Westview Press, Boulder, Colorado.

Vander Meer, R. K. V., S. D. Porter. 2001. Fate of newly mated queens introduced into monogyne and polygyne Solenopsis invicta (Hymenoptera : Formicidae) colonies. Annals of the Entomological Society of America 94: 289-297.

Vargo, E. L., C. Husseneder. 2011. 'Genetic structure of termite colonies and populations ' in 'Biology of Termites: A Modern Synthesis', pp. 321-348, edited by D. E. Bignell, Y. Roisin, and N. Lo. Springer, New York.

Vasquez, G. M., C. Schal, J. Silverman. 2009. Colony Fusion in Argentine Ants is Guided by Worker and Queen Cuticular Hydrocarbon Profile Similarity. Journal of Chemical

Ecology 35: 922-932. Vasquez, G. M., J. Silverman. 2008. Intraspecific aggression and colony fusion in the

Argentine ant. Animal Behaviour 75: 583-593. Verschuren, D., T. C. Johnson, H. J. Kling, D. N. Edgington, P. R. Leavitt, E. T. Brown, M.

R. Talbot, R. E. Hecky. 2002. History and timing of human impact on Lake Victoria, East Africa. Proceedings of the Royal Society B-Biological Sciences 269: 289-294.

Vitousek, P. M., C. M. DAntonio, L. L. Loope, R. Westbrooks. 1996. Biological invasions as global environmental change. American Scientist 84: 468-478.

Vogel, V., J. S. Pedersen, P. d'Ettorre, L. Lehmann, L. Keller. 2009. Dynamics and genetic structure of Argentine Ant supercolonies in their native range. Evolution 63: 1627-1639.

Vogel, V., J. S. Pedersen, T. Giraud, M. J. B. Krieger, L. Keller. 2010. The worldwide expansion of the Argentine ant. Diversity and Distributions 16: 170-186.

Walters, A. C., D. A. Mackay. 2005. Importance of large colony size for successful invasion by Argentine ants (Hymenoptera : Formicidae): Evidence for biotic resistance by native ants. Austral Ecology 30: 395-406.

X. References

114

Wanger, T. C., A. C. Wielgoss, I. Motzke, Y. Clough, B. B. W., N. S. Sodhi, T. Tscharnke. in press. Endemic predators, invasive prey and native diversity. Proceedings of the Royal

Society B - Biological Sciences. Ward, D. F., J. R. Beggs, M. N. Clout, R. J. Harris, S. O'Connor. 2006. The diversity and

origin of exotic ants arriving in New Zealand via human-mediated dispersal. Diversity

and Distributions 12: 601-609. Weeks, R. D., L. T. Wilson, S. B. Vinson, W. D. James. 2004. Flow of carbohydrates, lipids,

and protein among colonies of polygyne red imported fire ants, Solenopsis invicta (Hymenoptera : Formicidae). Annals of the Entomological Society of America 97: 105-110.

Wetterer, J. K. 2005. Worldwide distribution and potential spread of the long-legged Ant, Anoplolepis gracilipes (Hymenoptera: Formicidae). Sociobiology 45: 1-21.

Wetterer, J. K., J. A. Moore. 2005. Red imported fire ants (Hymenoptera : Formicidae) at gopher tortoise (Testudines : Testudinidae) burrows. Florida Entomologist 88: 349-354.

White, E. M., N. M. Sims, A. R. Clarke. 2008. Test of the enemy release hypothesis: The native magpie moth prefers a native fireweed (Senecio pinnatifolius) to its introduced congener (S. madagascariensis). Austral Ecology 33: 110-116.

Whitmore, T. C., J. A. Sayer. 1992. 'Deforestation and species extinction in tropical moist forests' in 'Tropical Deforestation and Species Extinction', pp. 1-12, edited by T. C. Whitmore, and J. A. Sayer. Chapman & Hall, London.

Williamson, M. 1996. Biological Invasions. Chapman & Hall, London. Williamson, M., A. Fitter. 1996. The varying success of invaders. Ecology 77: 1661-1666. Wilson, E. O. 1971. The Insect Societies. The Belknap Press of Harvard University Press,

Cambridge, Mass. Witte, F., B. S. Msuku, J. H. Wanink, O. Seehausen, E. F. B. Katunzi, P. C. Goudswaard, T.

Goldschmidt. 2000. Recovery of cichlid species in Lake Victoria: an examination of factors leading to differential extinction. Reviews in Fish Biology and Fisheries 10: 233-241.

Worm, B., E. B. Barbier, N. Beaumont, J. E. Duffy, C. Folke, B. S. Halpern, J. B. C. Jackson, H. K. Lotze, F. Micheli, S. R. Palumbi, E. Sala, K. A. Selkoe, J. J. Stachowicz, R. Watson. 2006. Impacts of biodiversity loss on ocean ecosystem services. Science 314: 787-790.

XI. Acknowledgements

115

XI. ACKNOWLEDGEMENTS

During the years that I worked on this thesis, I experienced the aid and support of many

people. With these acknowledgements, I’d like to send my kindest regards to all of the people

that helped making this thesis possible.

Nico Blüthgen introduced me to the world of myrmecology and supervised both my Diploma

thesis and now this PhD thesis. Thanks for your time, your guidance, your patience as well as

your lectures in statistics. The entire work group (Insect-Plant-Interactions) is thanked for

inspiring work group meetings, which, besides enthusiastic scientific discussions, contained a

good deal of humour.

Heike Feldhaar introduced me to the world of population genetics and evolutionary biology.

Without your enthusiastic and benevolent criticism, I still wouldn’t know how to write a

decent manuscript or put together a compelling talk.

Thomas Schmitt introduced me to the immense complexity of chemical communication in

social insects, which is something that I would surely like to understand way better than I do

now. Thank you for agreeing to co-supervise this thesis.

Maja Babis, Jana Bühler, Janina Eulenburg, Stephan Rinke and Deborah Schweinfest

were students that I had the honour to supervise and which produced valuable results that

proved to be vital for my research. Thanks for your effort and the good time, both here in

Würzburg and in Borneo!

Karl-Eduard Linsenmair, Dieter Mahsberg and Brigitte Fiala are generally thanked for

their passionate involvement in the Department of Animal Ecology & Tropical Biology as

well as for valuable suggestions and discussions.

Wolfgang Rössler and Flavio Roces let me set up more or less permanent camp in the

facilities of the Department of Behavioural Physiology & Sociobiology, even when status as

guest scientist had long run out.

Birgit Bünger, Susanne Linke and Irina Stahl provided help with all kinds of paperwork,

envelopes, fax machines, bills, registrations and bookings as well as a good chat every once in

a while.

XI. Acknowledgements

116

Adrienne Gerber-Kurz, Annete Laudahn and Karin Möller taught me a great deal about

how to keep ants and how to deal with microsatellites.

I would like to thank the entire staff and especially the PhD students of the Departments of

Animal Ecology & Tropical Biology and Behavioural Physiology & Sociobiology for helpful

discussions, assistance, criticism, suggestions and an overall great time.

Datin Maryati Mohamed, Maklarin bin Lakim, Arthur Chung and Sylvia Alsistu

facilitated my field work in Borneo and are thanked for the numerous times that they agreed

to help me with my official matters.

Justinah Parantis and all the members of the Butterfly farm in Poring were of great help in

Poring by letting me use large parts of the Butterfly Farm for my experiments.

Kamin Gebok was of immense help in Poring, not only by being a generous landlord or

knowing people and making things happen, but also by letting me live in the most

comfortable bamboo houses I’ve ever lived in.

Eugene Tan and the entire staff of Uncle Tan’s offered invaluable assistance during my

studies, including accommodation, hardware and a hut to set up a field laboratory. Not to

mention an unforgettable jungle camp in the last remnants of the Kinabatangan sweetwater

swamp forest, where there’s much more to do than ‘just’ watch Bornean wildlife. HBS!

The following people are thanked for inspiring scientific communication and/or collaboration:

Kirsti Abbott, Serge Aron, Petr Baňař, Catherine Berney, Andrew Cline, Dennis

O’Dowd, Kai Dworschak, Heikki Helanterä, Sigfrid Ingrisch, Milan Janda, Robert

Junker, Laurent Keller, Petr Klimes, Alain Lenoir, Christine La Mendola, Florian

Menzel, Juliane Mooz, Shyama Pagad, Floyd W. Shockley, Andrew V. Suarez, Pavel

Štys and Darren Ward.

Last but not least, I’d like to thank my entire family and my friends for believing in me and

always supporting my passion for the natural world.

Those that I may have inadvertently forgotten to mention are hereby thanked for their help as

well as their understanding and forgiveness.

Publikationen

Feldhaar H, Drescher J, Blüthgen N (2006). Characterization of microsatellite primers for the invasive ant species Anoplolepis gracilipes. Molecular Ecology Notes 6: 912-914

Drescher J, Blüthgen N, Feldhaar H (2007). Population structure and intraspecific aggression in the invasive ant species Anoplolepis gracilipes in Malaysian Borneo. Molecular Ecology 16: 1453-1465

Drescher J, Blüthgen N, Schmitt T, Bühler J, Feldhaar H. (2010). Societies drifting apart? Behavioural, genetic and chemical differentiation between neighboring supercolonies in the yellow crazy ant Anoplolepis gracilipes. PLoS ONE 5(10): e13581.

Stys P, Banar P, Drescher J (2010). A new Oncylocotis (Hemiptera: Heteroptera: Enicocephalidae) from Sabah: A predacious species associated with the Yellow Crazy Ant, Anoplolepis gracilipes (Hymenoptera: Formicidae). Zootaxa 2651: 27-42

Drescher J, Feldhaar H, Blüthgen N (2011). Interspecific Aggression and Resource Monopolization of the invasive ant Anoplolepis gracilipes in Malaysan Borneo. Biotropica

43: 93-99

Documents

The Ecology and Population structure of the invasive ... · aufgefunden (mehrheitlich neue, unbeschriebene Arten), von denen die Grille Myrmecophilus pallidithorax eingehender untersucht